AIRWiki - User contributions [en]

Predictive BCI Speller based on Motor Imagery

2009-07-22T10:21:37Z

TizianoDalbis: /* Part 4: Documents */

== '''Part 1: Project profile''' ==

=== Project name ===
A predictive speller for a brain-computer interface based on motor imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)
* [[User:RobertoTedesco | Roberto Tedesco]] (researcher)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects affected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on specific brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into different choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents significant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors affect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

[[Image:Bci_gui.png]]

[[Image:Bci_feedback.png]]

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed
* 21/07/2009: thesis presentation!

== '''Part 4: Documents''' ==

* Thesis document: [http://www.slideshare.net/tizyweb/a-predictive-speller-for-a-braincomputer-interface-based-on-motorimagery-1752428 A predictive speller for a brain-computer interface based on motor imagery] [En]
* Presentation [http://www.slideshare.net/tizyweb/a-predictive-speller-for-a-braincomputer-interface-based-on-motorimagery A predictive speller for a brain-computer interface based on motor imagery (presentation)][En]
* Video [http://www.youtube.com/watch?v=R-tNE-y2QU0 A predictive speller for a brain-computer interface based on motor imagery (video)]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-07-22T10:03:42Z

TizianoDalbis: /* Part 4: Documents */

== '''Part 1: Project profile''' ==

=== Project name ===
A predictive speller for a brain-computer interface based on motor imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)
* [[User:RobertoTedesco | Roberto Tedesco]] (researcher)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects affected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on specific brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into different choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents significant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors affect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

[[Image:Bci_gui.png]]

[[Image:Bci_feedback.png]]

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed
* 21/07/2009: thesis presentation!

== '''Part 4: Documents''' ==

* Thesis document: [http://www.box.net/shared/q44iirv3n2 A predictive speller for a brain-computer interface based on motor imagery] [En]
* Presentation [http://www.slideshare.net/tizyweb/a-predictive-speller-for-a-braincomputer-interface-based-on-motorimagery A predictive speller for a brain-computer interface based on motor imagery (presentation)][En]
* Video [http://www.youtube.com/watch?v=R-tNE-y2QU0 A predictive speller for a brain-computer interface based on motor imagery (video)]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-07-22T10:02:37Z

TizianoDalbis: /* Part 2: Project description */

== '''Part 1: Project profile''' ==

=== Project name ===
A predictive speller for a brain-computer interface based on motor imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)
* [[User:RobertoTedesco | Roberto Tedesco]] (researcher)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects affected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on specific brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into different choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents significant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors affect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

[[Image:Bci_gui.png]]

[[Image:Bci_feedback.png]]

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed
* 21/07/2009: thesis presentation!

== '''Part 4: Documents''' ==

* Thesis document: [http://www.box.net/shared/q44iirv3n2 A predictive speller for a brain-computer interface based on motor imagery] [En]
* Presentation [http://www.slideshare.net/tizyweb/a-predictive-speller-for-a-braincomputer-interface-based-on-motorimagery A predictive speller for a brain-computer interface based on motor imagery (presentation)][En]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-07-22T10:02:21Z

TizianoDalbis: /* Part 4: Documents */

== '''Part 1: Project profile''' ==

=== Project name ===
A predictive speller for a brain-computer interface based on motor imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)
* [[User:RobertoTedesco | Roberto Tedesco]] (researcher)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects affected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on specific brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into different choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents significant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors affect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

[http://www.youtube.com/watch?v=R-tNE-y2QU0 Video]

[[Image:Bci_gui.png]]

[[Image:Bci_feedback.png]]

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed
* 21/07/2009: thesis presentation!

== '''Part 4: Documents''' ==

* Thesis document: [http://www.box.net/shared/q44iirv3n2 A predictive speller for a brain-computer interface based on motor imagery] [En]
* Presentation [http://www.slideshare.net/tizyweb/a-predictive-speller-for-a-braincomputer-interface-based-on-motorimagery A predictive speller for a brain-computer interface based on motor imagery (presentation)][En]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-07-22T09:56:48Z

TizianoDalbis: /* Part 3: Project tracking */

== '''Part 1: Project profile''' ==

=== Project name ===
A predictive speller for a brain-computer interface based on motor imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)
* [[User:RobertoTedesco | Roberto Tedesco]] (researcher)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects affected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on specific brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into different choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents significant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors affect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

[http://www.youtube.com/watch?v=R-tNE-y2QU0 Video]

[[Image:Bci_gui.png]]

[[Image:Bci_feedback.png]]

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed
* 21/07/2009: thesis presentation!

== '''Part 4: Documents''' ==

* Thesis document: [http://www.box.net/shared/q44iirv3n2 A predictive speller for a brain-computer interface based on motor imagery] [En]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-06-22T18:30:19Z

TizianoDalbis: /* Project name */

== '''Part 1: Project profile''' ==

=== Project name ===
A predictive speller for a brain-computer interface based on motor imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)
* [[User:RobertoTedesco | Roberto Tedesco]] (researcher)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects affected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on specific brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into different choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents significant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors affect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

[http://www.youtube.com/watch?v=R-tNE-y2QU0 Video]

[[Image:Bci_gui.png]]

[[Image:Bci_feedback.png]]

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed

== '''Part 4: Documents''' ==

* Thesis document: [http://www.box.net/shared/q44iirv3n2 A predictive speller for a brain-computer interface based on motor imagery] [En]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-06-22T18:26:01Z

TizianoDalbis: /* Part 4: Documents */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)
* [[User:RobertoTedesco | Roberto Tedesco]] (researcher)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects affected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on specific brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into different choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents significant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors affect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

[http://www.youtube.com/watch?v=R-tNE-y2QU0 Video]

[[Image:Bci_gui.png]]

[[Image:Bci_feedback.png]]

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed

== '''Part 4: Documents''' ==

* Thesis document: [http://www.box.net/shared/q44iirv3n2 A predictive speller for a brain-computer interface based on motor imagery] [En]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-06-22T12:57:10Z

TizianoDalbis: /* Part 2: Project description */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)
* [[User:RobertoTedesco | Roberto Tedesco]] (researcher)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects affected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on specific brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into different choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents significant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors affect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

[http://www.youtube.com/watch?v=R-tNE-y2QU0 Video]

[[Image:Bci_gui.png]]

[[Image:Bci_feedback.png]]

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed

== '''Part 4: Documents''' ==

* Thesis document: [[image:Thesis_predictive_bci_speller.pdf‎ | Thesis_predictive_bci_speller.pdf‎ ]] [En]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-06-22T12:40:11Z

TizianoDalbis: /* Part 2: Project description */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)
* [[User:RobertoTedesco | Roberto Tedesco]] (researcher)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects affected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on specific brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into different choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents significant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors affect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

[[Image:Bci_gui.png]]

[[Image:Bci_feedback.png]]

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed

== '''Part 4: Documents''' ==

* Thesis document: [[image:Thesis_predictive_bci_speller.pdf‎ | Thesis_predictive_bci_speller.pdf‎ ]] [En]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

File:Bci gui.png

2009-06-22T12:29:52Z

TizianoDalbis: Predictive BCI Speller based on Motor Imagery: Spelling application

Predictive BCI Speller based on Motor Imagery: Spelling application

File:Bci feedback.png

2009-06-22T12:28:57Z

TizianoDalbis: Predictive BCI Speller Based on Motor Imagery: feedback sessions

Predictive BCI Speller Based on Motor Imagery: feedback sessions

User:TizianoDalbis

2009-06-19T13:09:41Z

TizianoDalbis:

{{SMWUser
|firstname=Tiziano
|lastname=D'Albis
|email=tiziano(dot)dalbis(at)gmail(dot)com
|advisor=MatteoMatteucci
|projectpage=BCI based on Motor Imagery
|photo=TizianoDalbis.png
}}

I am a Master Student in Computer Engineering at Politecnico di Milano (Como campus). 
I'm currently working on my master thesis about [[Predictive_BCI_Speller_based_on_Motor_Imagery | Predictive BCI Speller based on motor imagery]] with [[User:MatteoMatteucci | Matteo Matteucci]] and [[User:LiciaSbattella | Licia Sbattella]] . 

My interests span from software and web engineering to AI and philosophy. 
During the spare time I like walking in mountain, cycling, taking pictures and of course enjoying with my friends! 
I'm also the secretary and founding member of the cutural youth association "Tricheco" in my town.

Here are some some links you may want to follow:
* My curriculum vitae http://www.box.net/shared/8a9bg6qkg0
* My Facebook profile http://www.facebook.com/profile.php?id=778283933
* My gallery on DeviantArt http://tizyweb.deviantart.com
* My photo gallery on PicasaWeb http://picasaweb.google.it/tizyweb
* My library on Anobii http://www.anobii.com/people/tizyweb/
* The website of my association: http://www.tricheco.altervista.org

Predictive BCI Speller based on Motor Imagery

2009-06-19T13:06:47Z

TizianoDalbis: /* Part 4: Documents */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)
* [[User:RobertoTedesco | Roberto Tedesco]] (researcher)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects affected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on specific brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into different choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents significant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors affect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed

== '''Part 4: Documents''' ==

* Thesis document: [[image:Thesis_predictive_bci_speller.pdf‎ | Thesis_predictive_bci_speller.pdf‎ ]] [En]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-06-19T13:06:30Z

TizianoDalbis: /* Part 4: Documents */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)
* [[User:RobertoTedesco | Roberto Tedesco]] (researcher)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects affected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on specific brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into different choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents significant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors affect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed

== '''Part 4: Documents''' ==

* Thesis description document: [[image:Thesis_predictive_bci_speller.pdf‎ | Thesis_predictive_bci_speller.pdf‎ ]] [En]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

File:Thesis predictive bci speller.pdf

2009-06-19T13:04:12Z

TizianoDalbis: Master thesis: A predictive BCI speller for a brain-computer interface based on motor imagery

Master thesis: A predictive BCI speller for a brain-computer interface based on motor imagery

Predictive BCI Speller based on Motor Imagery

2009-06-19T13:01:21Z

TizianoDalbis: /* Other Politecnico di Milano people */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)
* [[User:RobertoTedesco | Roberto Tedesco]] (researcher)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects affected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on specific brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into different choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents significant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors affect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed

== '''Part 4: Documents''' ==

* Thesis description document: [[image:predictive_BCI_speller_v1.pdf | predictive_BCI_speller_v1.pdf ]] [ITA] [Draft] [Last update: 11/11/2008]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-06-19T13:00:29Z

TizianoDalbis: /* Part 2: Project description */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects affected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on specific brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into different choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents significant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors affect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed

== '''Part 4: Documents''' ==

* Thesis description document: [[image:predictive_BCI_speller_v1.pdf | predictive_BCI_speller_v1.pdf ]] [ITA] [Draft] [Last update: 11/11/2008]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-06-19T12:59:22Z

TizianoDalbis: /* Part 2: Project description */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The objective of this thesis is to implement a BCI spelling application for people with severe motor impairments, adopting natural language processing techniques to improve the overall communication rate of the system. The target users of this application are subjects aﬀected by severe motor disorders and in particular patients affected by the locked-in syndrome. Communication is performed detecting different mental states from the subject’s EEG and translating these states into input commands for a spelling application.

The BCI paradigm adopted is motor imagery. When the subject imagines to move a certain part of the body (such as a hand or a foot) produces modiﬁcations on speciﬁc brain rhythms characteristic of the motor cortex. These modiﬁcations are detected in real-time from the EEG signal and mapped into diﬀerent choices in the spelling application. This kind of communication is typically very slow and prone to errors. The EEG signal recorded from the scalp has indeed a low signal-to-noise ratio, can be affected by a number of artifacts and presents signiﬁcant variability with different acquisition settings. Moreover the neurophysiological phenomena related to motor imagery manifest differently with different subjects and may also be influenced by the particular psychophysical conditions met during the task.

All these factors aﬀect the performances of the brain-computer interface.
The control signal received by the spelling application is indeed characterized by low information transfer rates and low accuracy levels. This is the major problem that this thesis had to face, being the final objective of this work to obtain a BCI system suitable for effective verbal communication. The approach chosen in order to maximize the overall communication rate is two-fold: on one side there is an effort in gaining information transfer rate from the control signal, on the other side there is an effort in optimizing the way this information is employed for the purpose of verbal communication.

Dealing with the first problem, this thesis focuses on the design of novel features extracted from the EEG signal in the frequency domain and on the adoption of different machine learning techniques for feature selection and classification. Dealing with the second problem, instead, we designed an original speller interface in which redundancies in natural language are used to speed up the selection of symbols and word suggestions are provided to the user during the composition.

The spelling application developed is thus composed of three main modules: the brain-computer interface, the speller interface and the language prediction module. Each module has been designed, implemented and tested specifically for the objectives of this thesis and the results have been evaluated at different levels. First we assessed the performances of the BCI module alone, both offline (with pre-recorded signals) and online (with signals acquired and classified in real-time). Then we evaluated the performances of the spelling interface with a simulator program (implemented for the purpose) and we assessed the impact of classification accuracy and language predictions on the overall system performances. Finally we tested the whole BCI spelling application online considering different subjects and different acquisition sessions.

The achieved results are very satisfactory and comparable with the latest works about motor-imagery BCI spellers reported in literature. With one subject we obtained high classification accuracies and the overall communication rate achieved is 3 char/min. With other two subjects, instead, we started with lower classification accuracies, but significant improvements have been obtained as the number of training sessions increased. These subjects reached a spelling speed of respectively 2 char/min and 2.7 char/min.

A trend of performance improvement has been observed in our tests and we believe that with more feedback sessions an even better BCI control could be obtained.
Further improvements may be also achieved considering EEG features in the spatial domain (for example with the method of Common Spatial Patterns) and testing other classification algorithms. Moreover alternative symbol arrangements may be investigated in order to speedup the selection process and the accuracy of language predictions could be improved by means of customized language models.

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed

== '''Part 4: Documents''' ==

* Thesis description document: [[image:predictive_BCI_speller_v1.pdf | predictive_BCI_speller_v1.pdf ]] [ITA] [Draft] [Last update: 11/11/2008]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-06-19T12:52:55Z

TizianoDalbis: /* Part 3: Project tracking */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The goal of this project is to implement a BCI system supporting people with motor disabilities to communicate effectively.
The main idea is to adopt the "virtual keyboard" paradigm in which a series of symbols are selected by means of a BCI interface.

This kind of communication can be included in the class of Alternative and Augmentative Communication (AAC) and the main issue with it is in the need of minimizing the number of selections required by the user while running the application. Indeed the selection process could be costly (both in terms of time and effort) for the target users of these kinds of systems.

The adoption of a BCI interface, moreover, sets also a limit to the number of alternative choices available for each selection. This limit depends on the specific BCI paradigm used and on the way the selection process is carried out. Using a scanning approach, for example, it is possible to overcome this kind of limit, but introducing the drawback of much longer selection delays.

On the other side with a direct selection it is possible to classify effectively a very limited number of states, with common BCI paradigms this limit is usually set to 2,3 or 4.
The Motor Imagery paradigm, instead, is somehow different because it could be used to translate sensymotor rhythms in the continuous movement of a cursor in 2 dimensions. In this case it is possible to associate a choice with a target on the screen and perform a selection just hitting a screen target with the cursor. With this kind of approach the number of different choices is limited only by the precision of the cursor movement that can be achieved.

The main idea of this thesis is group the alphabetical letters in sets and than recursively expand them (with a selection based on motor imagery) till a single letter is selected. Moreover we want to aid the selection process displaying the most probable letters in sets with a low number of elements, thus minimizing the selection steps required by the user. Finally we want to implement also a word completion and prediction algorithm that would give also the possibility to select a whole word in one shot.

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis
* 09/04/2009: started testing the spelling application
* 18/06/2009: thesis document completed

== '''Part 4: Documents''' ==

* Thesis description document: [[image:predictive_BCI_speller_v1.pdf | predictive_BCI_speller_v1.pdf ]] [ITA] [Draft] [Last update: 11/11/2008]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

IIT-Lab

2009-06-04T12:30:22Z

TizianoDalbis: /* Booking */

== What is the IIT-Lab ==

AIRLab-IITLab is dedicated to activities founded by the Italian Institute of Technology.
The lab hosts activities related to Brain-Computer Interfaces (BCI) and Affective Computing.

=== Location ===
It is located in the Rimembranze di Lambrate building of the Department of Electronics and Information, Via Rimembranze di Lambrate, 14, Milan.

'''The Lab has moved. It is now located at the first floor just over the Airlab, always in the Rimembranze di Lambrate building.'''

=== Access Rules ===
The access to AIRLab-IITLab is reserved to registered users. If you are student and want to register, you have to fill the AIRLab registration form (to be signed by your tutor) and the security form. The key of the lab is provided to registered users by the doorkeeper at the main entrance of the Lambrate building.

=== Booking ===

Please book the instrument you want to use by adding an entry to the table; the booking of an instrument implies the booking of the room. If you want to use a different instrument at the same time of an existing booking, please contact the other person involved and check that you can share the room; alternatively, you can ask the doorkeeper for an empty room in the building.


{| border="1"
! Day !! Time !! Person !! Instrument
|-
| Friday 5 June|| 13:00-18:30 || [[User:FrancescoSpadoni]] || RehabRobot
|-
|}

== Links ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Affective Computing]] page on this Wiki
* [http://www.airlab.elet.polimi.it/index.php/airlab/visitor_info/airlab_iitlab AIRLab - IITLab]

IIT-Lab

2009-06-02T16:24:52Z

TizianoDalbis: /* Booking */

== What is the IIT-Lab ==

AIRLab-IITLab is dedicated to activities founded by the Italian Institute of Technology.
The lab hosts activities related to Brain-Computer Interfaces (BCI) and Affective Computing.

=== Location ===
It is located in the Rimembranze di Lambrate building of the Department of Electronics and Information, Via Rimembranze di Lambrate, 14, Milan.

'''The Lab has moved. It is now located at the first floor just over the Airlab, always in the Rimembranze di Lambrate building.'''

=== Access Rules ===
The access to AIRLab-IITLab is reserved to registered users. If you are student and want to register, you have to fill the AIRLab registration form (to be signed by your tutor) and the security form. The key of the lab is provided to registered users by the doorkeeper at the main entrance of the Lambrate building.

=== Booking ===

Please book the instrument you want to use by adding an entry to the table; the booking of an instrument implies the booking of the room. If you want to use a different instrument at the same time of an existing booking, please contact the other person involved and check that you can share the room; alternatively, you can ask the doorkeeper for an empty room in the building.


{| border="1"
! Day !! Time !! Person !! Instrument
|-
| Wednesday 3 June|| 9:00-19:30 || [[User:TizianoDalbis]] || EEG
|-
|}

== Links ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Affective Computing]] page on this Wiki
* [http://www.airlab.elet.polimi.it/index.php/airlab/visitor_info/airlab_iitlab AIRLab - IITLab]

IIT-Lab

2009-05-24T21:49:56Z

TizianoDalbis: /* Booking */

== What is the IIT-Lab ==

AIRLab-IITLab is dedicated to activities founded by the Italian Institute of Technology.
The lab hosts activities related to Brain-Computer Interfaces (BCI) and Affective Computing.

=== Location ===
It is located in the Rimembranze di Lambrate building of the Department of Electronics and Information, Via Rimembranze di Lambrate, 14, Milan.

'''The Lab has moved. It is now located at the first floor just over the Airlab, always in the Rimembranze di Lambrate building.'''

=== Access Rules ===
The access to AIRLab-IITLab is reserved to registered users. If you are student and want to register, you have to fill the AIRLab registration form (to be signed by your tutor) and the security form. The key of the lab is provided to registered users by the doorkeeper at the main entrance of the Lambrate building.

=== Booking ===

Please book the instrument you want to use by adding an entry to the table; the booking of an instrument implies the booking of the room. If you want to use a different instrument at the same time of an existing booking, please contact the other person involved and check that you can share the room; alternatively, you can ask the doorkeeper for an empty room in the building.


{| border="1"
! Day !! Time !! Person !! Instrument
|-
| Monday 25 May|| 9:00-19:30 || [[User:TizianoDalbis]] || EEG
|-
|}

== Links ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Affective Computing]] page on this Wiki
* [http://www.airlab.elet.polimi.it/index.php/airlab/visitor_info/airlab_iitlab AIRLab - IITLab]

IIT-Lab

2009-05-18T15:28:49Z

TizianoDalbis: /* Booking */

== What is the IIT-Lab ==

AIRLab-IITLab is dedicated to activities founded by the Italian Institute of Technology.
The lab hosts activities related to Brain-Computer Interfaces (BCI) and Affective Computing.

=== Location ===
It is located in the Rimembranze di Lambrate building of the Department of Electronics and Information, Via Rimembranze di Lambrate, 14, Milan.

'''The Lab has moved. It is now located at the first floor just over the Airlab, always in the Rimembranze di Lambrate building.'''

=== Access Rules ===
The access to AIRLab-IITLab is reserved to registered users. If you are student and want to register, you have to fill the AIRLab registration form (to be signed by your tutor) and the security form. The key of the lab is provided to registered users by the doorkeeper at the main entrance of the Lambrate building.

=== Booking ===

Please book the instrument you want to use by adding an entry to the table; the booking of an instrument implies the booking of the room. If you want to use a different instrument at the same time of an existing booking, please contact the other person involved and check that you can share the room; alternatively, you can ask the doorkeeper for an empty room in the building.


{| border="1"
! Day !! Time !! Person !! Instrument
|-
| Every wednesday || 9:00-19:30 || [[User:RossellaBlatt]] || EEG
|-
| Every thursday || 9:00-19:30 || [[User:RossellaBlatt]] || EEG
|-
| Tuesday 19 May|| 9:00-19:30 || [[User:TizianoDabis]] || EEG
|-
|}

== Links ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Affective Computing]] page on this Wiki
* [http://www.airlab.elet.polimi.it/index.php/airlab/visitor_info/airlab_iitlab AIRLab - IITLab]

IIT-Lab

2009-05-18T15:28:26Z

TizianoDalbis: /* Booking */

== What is the IIT-Lab ==

AIRLab-IITLab is dedicated to activities founded by the Italian Institute of Technology.
The lab hosts activities related to Brain-Computer Interfaces (BCI) and Affective Computing.

=== Location ===
It is located in the Rimembranze di Lambrate building of the Department of Electronics and Information, Via Rimembranze di Lambrate, 14, Milan.

'''The Lab has moved. It is now located at the first floor just over the Airlab, always in the Rimembranze di Lambrate building.'''

=== Access Rules ===
The access to AIRLab-IITLab is reserved to registered users. If you are student and want to register, you have to fill the AIRLab registration form (to be signed by your tutor) and the security form. The key of the lab is provided to registered users by the doorkeeper at the main entrance of the Lambrate building.

=== Booking ===

Please book the instrument you want to use by adding an entry to the table; the booking of an instrument implies the booking of the room. If you want to use a different instrument at the same time of an existing booking, please contact the other person involved and check that you can share the room; alternatively, you can ask the doorkeeper for an empty room in the building.


{| border="1"
! Day !! Time !! Person !! Instrument
|-
| Every wednesday || 9:00-19:30 || [[User:RossellaBlatt]] || EEG
|-
| Every thursday || 9:00-19:30 || [[User:RossellaBlatt]] || EEG
|-
| Tuesday 19 Maj|| 9:00-19:30 || [[User:TizianoDabis]] || EEG
|-
|}

== Links ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Affective Computing]] page on this Wiki
* [http://www.airlab.elet.polimi.it/index.php/airlab/visitor_info/airlab_iitlab AIRLab - IITLab]

IIT-Lab

2009-04-25T07:44:16Z

TizianoDalbis: /* Booking */

== What is the IIT-Lab ==

AIRLab-IITLab is dedicated to activities founded by the Italian Institute of Technology.
The lab hosts activities related to Brain-Computer Interfaces (BCI) and Affective Computing.

=== Location ===
It is located in the Rimembranze di Lambrate building of the Department of Electronics and Information, Via Rimembranze di Lambrate, 14, Milan.

=== Access Rules ===
The access to AIRLab-IITLab is reserved to registered users. If you are student and want to register, you have to fill the AIRLab registration form (to be signed by your tutor) and the security form. The key of the lab is provided to registered users by the doorkeeper at the main entrance of the Lambrate building.

=== Booking ===

Please book the instrument you want to use by adding an entry to the table; the booking of an instrument implies the booking of the room. If you want to use a different instrument at the same time of an existing booking, please contact the other person involved and check that you can share the room; alternatively, you can ask the doorkeeper for an empty room in the building.


{| border="1"
! Day !! Time !! Person !! Instrument
|-
| Every wednesday || 9:00-19:30 || [[User:RossellaBlatt]] || EEG
|-
| Every thursday || 9:00-19:30 || [[User:RossellaBlatt]] || EEG
|-
| Friday 24 April || 14:30-18:00 || [[User:FedericoRinaldi]] [[User:FrancescoSpadoni]] || RehabRobot
|-
| Monday 29 April || 9:00-19:30 || [[User:TizianoDalbis]] || EEG
|-
|}

== Links ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Affective Computing]] page on this Wiki
* [http://www.airlab.elet.polimi.it/index.php/airlab/visitor_info/airlab_iitlab AIRLab - IITLab]

Predictive BCI Speller based on Motor Imagery

2009-04-14T19:03:43Z

TizianoDalbis: /* Part 3: Project tracking */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The goal of this project is to implement a BCI system supporting people with motor disabilities to communicate effectively.
The main idea is to adopt the "virtual keyboard" paradigm in which a series of symbols are selected by means of a BCI interface.

This kind of communication can be included in the class of Alternative and Augmentative Communication (AAC) and the main issue with it is in the need of minimizing the number of selections required by the user while running the application. Indeed the selection process could be costly (both in terms of time and effort) for the target users of these kinds of systems.

The adoption of a BCI interface, moreover, sets also a limit to the number of alternative choices available for each selection. This limit depends on the specific BCI paradigm used and on the way the selection process is carried out. Using a scanning approach, for example, it is possible to overcome this kind of limit, but introducing the drawback of much longer selection delays.

On the other side with a direct selection it is possible to classify effectively a very limited number of states, with common BCI paradigms this limit is usually set to 2,3 or 4.
The Motor Imagery paradigm, instead, is somehow different because it could be used to translate sensymotor rhythms in the continuous movement of a cursor in 2 dimensions. In this case it is possible to associate a choice with a target on the screen and perform a selection just hitting a screen target with the cursor. With this kind of approach the number of different choices is limited only by the precision of the cursor movement that can be achieved.

The main idea of this thesis is group the alphabetical letters in sets and than recursively expand them (with a selection based on motor imagery) till a single letter is selected. Moreover we want to aid the selection process displaying the most probable letters in sets with a low number of elements, thus minimizing the selection steps required by the user. Finally we want to implement also a word completion and prediction algorithm that would give also the possibility to select a whole word in one shot.

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000
* 09/04/2009: finished implemented the feedback application module in BCI2000
* 09/04/2009: started writing the thesis

== '''Part 4: Documents''' ==

* Thesis description document: [[image:predictive_BCI_speller_v1.pdf | predictive_BCI_speller_v1.pdf ]] [ITA] [Draft] [Last update: 11/11/2008]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-03-28T14:31:40Z

TizianoDalbis: /* Part 3: Project tracking */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The goal of this project is to implement a BCI system supporting people with motor disabilities to communicate effectively.
The main idea is to adopt the "virtual keyboard" paradigm in which a series of symbols are selected by means of a BCI interface.

This kind of communication can be included in the class of Alternative and Augmentative Communication (AAC) and the main issue with it is in the need of minimizing the number of selections required by the user while running the application. Indeed the selection process could be costly (both in terms of time and effort) for the target users of these kinds of systems.

The adoption of a BCI interface, moreover, sets also a limit to the number of alternative choices available for each selection. This limit depends on the specific BCI paradigm used and on the way the selection process is carried out. Using a scanning approach, for example, it is possible to overcome this kind of limit, but introducing the drawback of much longer selection delays.

On the other side with a direct selection it is possible to classify effectively a very limited number of states, with common BCI paradigms this limit is usually set to 2,3 or 4.
The Motor Imagery paradigm, instead, is somehow different because it could be used to translate sensymotor rhythms in the continuous movement of a cursor in 2 dimensions. In this case it is possible to associate a choice with a target on the screen and perform a selection just hitting a screen target with the cursor. With this kind of approach the number of different choices is limited only by the precision of the cursor movement that can be achieved.

The main idea of this thesis is group the alphabetical letters in sets and than recursively expand them (with a selection based on motor imagery) till a single letter is selected. Moreover we want to aid the selection process displaying the most probable letters in sets with a low number of elements, thus minimizing the selection steps required by the user. Finally we want to implement also a word completion and prediction algorithm that would give also the possibility to select a whole word in one shot.

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value
* 27/03/2009: finished implementing phase lock value, relevant correlations found
* 27/03/2009: started implementing the classification algorithm in BCI2000

== '''Part 4: Documents''' ==

* Thesis description document: [[image:predictive_BCI_speller_v1.pdf | predictive_BCI_speller_v1.pdf ]] [ITA] [Draft] [Last update: 11/11/2008]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-03-23T19:47:59Z

TizianoDalbis: /* Part 3: Project tracking */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The goal of this project is to implement a BCI system supporting people with motor disabilities to communicate effectively.
The main idea is to adopt the "virtual keyboard" paradigm in which a series of symbols are selected by means of a BCI interface.

This kind of communication can be included in the class of Alternative and Augmentative Communication (AAC) and the main issue with it is in the need of minimizing the number of selections required by the user while running the application. Indeed the selection process could be costly (both in terms of time and effort) for the target users of these kinds of systems.

The adoption of a BCI interface, moreover, sets also a limit to the number of alternative choices available for each selection. This limit depends on the specific BCI paradigm used and on the way the selection process is carried out. Using a scanning approach, for example, it is possible to overcome this kind of limit, but introducing the drawback of much longer selection delays.

On the other side with a direct selection it is possible to classify effectively a very limited number of states, with common BCI paradigms this limit is usually set to 2,3 or 4.
The Motor Imagery paradigm, instead, is somehow different because it could be used to translate sensymotor rhythms in the continuous movement of a cursor in 2 dimensions. In this case it is possible to associate a choice with a target on the screen and perform a selection just hitting a screen target with the cursor. With this kind of approach the number of different choices is limited only by the precision of the cursor movement that can be achieved.

The main idea of this thesis is group the alphabetical letters in sets and than recursively expand them (with a selection based on motor imagery) till a single letter is selected. Moreover we want to aid the selection process displaying the most probable letters in sets with a low number of elements, thus minimizing the selection steps required by the user. Finally we want to implement also a word completion and prediction algorithm that would give also the possibility to select a whole word in one shot.

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: Formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities
* 18/03/2009: working on feature selection for EEG classification
* 21/03/2009: implemented a genetic algorithm for feature selection
* 23/03/2009: working on phase lock value for feature selection

== '''Part 4: Documents''' ==

* Thesis description document: [[image:predictive_BCI_speller_v1.pdf | predictive_BCI_speller_v1.pdf ]] [ITA] [Draft] [Last update: 11/11/2008]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-03-12T19:49:13Z

TizianoDalbis: /* Part 3: Project tracking */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The goal of this project is to implement a BCI system supporting people with motor disabilities to communicate effectively.
The main idea is to adopt the "virtual keyboard" paradigm in which a series of symbols are selected by means of a BCI interface.

This kind of communication can be included in the class of Alternative and Augmentative Communication (AAC) and the main issue with it is in the need of minimizing the number of selections required by the user while running the application. Indeed the selection process could be costly (both in terms of time and effort) for the target users of these kinds of systems.

The adoption of a BCI interface, moreover, sets also a limit to the number of alternative choices available for each selection. This limit depends on the specific BCI paradigm used and on the way the selection process is carried out. Using a scanning approach, for example, it is possible to overcome this kind of limit, but introducing the drawback of much longer selection delays.

On the other side with a direct selection it is possible to classify effectively a very limited number of states, with common BCI paradigms this limit is usually set to 2,3 or 4.
The Motor Imagery paradigm, instead, is somehow different because it could be used to translate sensymotor rhythms in the continuous movement of a cursor in 2 dimensions. In this case it is possible to associate a choice with a target on the screen and perform a selection just hitting a screen target with the cursor. With this kind of approach the number of different choices is limited only by the precision of the cursor movement that can be achieved.

The main idea of this thesis is group the alphabetical letters in sets and than recursively expand them (with a selection based on motor imagery) till a single letter is selected. Moreover we want to aid the selection process displaying the most probable letters in sets with a low number of elements, thus minimizing the selection steps required by the user. Finally we want to implement also a word completion and prediction algorithm that would give also the possibility to select a whole word in one shot.

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm
* 12/03/2009: Formal description (with Push Down Automata) of two interfaces and simulation of their performances with different probabilities

== '''Part 4: Documents''' ==

* Thesis description document: [[image:predictive_BCI_speller_v1.pdf | predictive_BCI_speller_v1.pdf ]] [ITA] [Draft] [Last update: 11/11/2008]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-03-07T15:38:39Z

TizianoDalbis: /* Part 3: Project tracking */

== '''Part 1: Project profile''' ==

=== Project name ===

BCI based on Motor Imagery

=== Project short description ===

This project is aimed is to control an external device through the analysis of brain waves measured on the human scalp.

=== Dates ===
Start date: 01/05/2008

End date:

=== Website(s) ===

http://airlab.elet.polimi.it/index.php/airlab/theses_lab_projects/brain_computer_interfaces_based_on_motor_imagery

=== People involved ===

===== Project head(s) =====

* [[User:MatteoMatteucci | Matteo Matteucci]] (professor)
* [[User:SbattellaLicia | Sbattella Licia]] (professor)

===== Other Politecnico di Milano people =====

* [[User:RossellaBlatt | Rossella Blatt]] (phd student)

===== Students currently working on the project =====

* [[User:TizianoDalbis | Tiziano D'Albis]] (master student)

=== Laboratory work and risk analysis ===

Laboratory work for this project will be mainly performed at AIRLab-IIT/Lambrate. The main activity consists in the acquisition of brain signals through an EEG amplifier for on-line or off-line processing. This is a potentially risky activity since there is an electrical instrumentation that is in direct contact with the human body. It is thus important to keep the system isolated from the power line. The EEG amplifier (as all biomedical instrumentations) is certified by the vendor to be isolated and the acquired data are transferred to the PC using an optic fiber connection . Anyhow for increased safety the PC and any other electronic device connected to the system must be disconnected from the power line.
Standard safety measures described in [http://airlab.elet.polimi.it/index.php/airlab/content/download/461/4110/file/documento_valutazione_rischi_AIRLab.pdf Safety norms] will be followed.

== '''Part 2: Project description''' ==

The goal of this project is to implement a BCI system supporting people with motor disabilities to communicate effectively.
The main idea is to adopt the "virtual keyboard" paradigm in which a series of symbols are selected by means of a BCI interface.

This kind of communication can be included in the class of Alternative and Augmentative Communication (AAC) and the main issue with it is in the need of minimizing the number of selections required by the user while running the application. Indeed the selection process could be costly (both in terms of time and effort) for the target users of these kinds of systems.

The adoption of a BCI interface, moreover, sets also a limit to the number of alternative choices available for each selection. This limit depends on the specific BCI paradigm used and on the way the selection process is carried out. Using a scanning approach, for example, it is possible to overcome this kind of limit, but introducing the drawback of much longer selection delays.

On the other side with a direct selection it is possible to classify effectively a very limited number of states, with common BCI paradigms this limit is usually set to 2,3 or 4.
The Motor Imagery paradigm, instead, is somehow different because it could be used to translate sensymotor rhythms in the continuous movement of a cursor in 2 dimensions. In this case it is possible to associate a choice with a target on the screen and perform a selection just hitting a screen target with the cursor. With this kind of approach the number of different choices is limited only by the precision of the cursor movement that can be achieved.

The main idea of this thesis is group the alphabetical letters in sets and than recursively expand them (with a selection based on motor imagery) till a single letter is selected. Moreover we want to aid the selection process displaying the most probable letters in sets with a low number of elements, thus minimizing the selection steps required by the user. Finally we want to implement also a word completion and prediction algorithm that would give also the possibility to select a whole word in one shot.

== '''Part 3: Project tracking''' ==

* 01/04/2008: started studying literature about BCI and Motor Imagery
* 01/05/2008: started acquiring data with BCI2000 - performing initial sessions
* 01/06/2008: stated performing feedback sessions with cursor movement in 1D
* 01/09/2008: first idea about the predictive BCI speller
* 10/09/2008: started studying literature about AAC and NLP
* 24/09/2008: published the first document describing the thesis
* 21/10/2008: started studying the thesis about "Soothsayer" by Matteo Vescovi
* 21/10/2008: started studying how to implement in BCI2000 the algorithm of weights adaptation for 2D cursor movement
* 24/10/2008: finished studying the thesis about "Soothsayer" by Matteo Vescovi
* 24/10/2008: added a chapter in thesis description about the weights adaptation algorithm
* 31/10/2008: thesis description document update: added a detailed description of the BCI, new appendix about R^2 calculation
* 03/11/2008: thesis description document update: started a new chapter about letter prediction
* 03/11/2008: thesis description document update: updated the section on weights adaptation
* 05/12/2008: implemented two algorithms for weight adaptation: RLMS and Delta rule
* 10/12/2008: implemented a batch MATLAB script for the off-line analysis
* 15/12/2008: implemented a MATLAB script to simulate the on-line computation of the spectrum
* 20/12/2008: designed classes for the Speller application
* 02/01/2009: started implementing the Speller application in BCI2000
* 23/01/2009: first demo of the Speller controlled by keyboard and without letter prediction
* 05/02/2009: demo of the Speller with letter prediciton
* 05/02/2009: working on feature selection/extraction for the offline classification of the EEG
* 10/02/2009: first result data about classification accuracy inside the same session
* 10/02/2009: working on classification accuracy across sessions
* 15/02/2009: developing Speller application: formally description of the GUI behaviour using a FSM
* 20/02/2009: implemented XML configurability of the interface in the Speller application
* 22/02/2009: implemented a patch for Presage to customize prediction features
* 23/02/2009: generated n-gram statistics from a partition of the British National Corpus
* 25/02/2009: redesign of the interface structure, refactoring of the Speller application code, implementation of new Speller features
* 01/03/2009: working on the Speller simulator program
* 02/03/2009: first version of the Speller simulator without errors and with pre-calculated paths
* 07/03/2009: Speller simulator implemented as a stack machine, including error handling and a breadth-first path-finding algorithm

== '''Part 4: Documents''' ==

* Thesis description document: [[image:predictive_BCI_speller_v1.pdf | predictive_BCI_speller_v1.pdf ]] [ITA] [Draft] [Last update: 11/11/2008]

== '''Part 5: References''' ==

* Control of two-dimensional movement signals by a noninvasive brain-computer interface in humans, Wolpaw J.R., McFarland J., PNAS, vol. 101, no. 51, december 2004, pages 17849-17854.
* Brain Computer interfaces for communication and control, Wolpaw J.R., Birbaumer N., McFarland D., Pfurtsheller G., Vaughan T., Clinical Neurophysiology 113, 2002, 767-791
* EEG based communication: prospects and problems, Vaughan T., Wolpaw J.R., Donchin E., IEEE transactions on rehabilitation engineering, vol. 4, no. 4, december 1996, pages 425-430.

== '''Part 6: Links''' ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Electroencephalographs]]
* [[How to mount electrodes]]
* [[How to setup BCI software]]

Predictive BCI Speller based on Motor Imagery

2009-02-07T12:28:42Z

TizianoDalbis: /* Part 3: Project tracking */

IIT-Lab

2009-01-28T17:07:26Z

TizianoDalbis: /* Booking */

== What is the IIT-Lab ==

AIRLab-IITLab is dedicated to activities founded by the Italian Institute of Technology.
The lab hosts activities related to Brain-Computer Interfaces (BCI) and Affective Computing.

=== Location ===
It is located in the Rimembranze di Lambrate building of the Department of Electronics and Information, Via Rimembranze di Lambrate, 14, Milan.

=== Access Rules ===
The access to AIRLab-IITLab is reserved to registered users. If you are student and want to register, you have to fill the AIRLab registration form (to be signed by your tutor) and the security form. The key of the lab is provided to registered users by the doorkeeper at the main entrance of the Lambrate building.

=== Booking ===

Please book the instrument you want to use by adding an entry to the table; the booking of an instrument implies the booking of the room. If you want to use a different instrument at the same time of an existing booking, please contact the other person involved and check that you can share the room; alternatively, you can ask the doorkeeper for an empty room in the building.


{| border="1"
! Day !! Time !! Person !! Instrument
|-
|29 Jan || 9:30-18:00 || [[User:TizianoDalbis]] || EEG
|-
|30 Jan || 9:30-18:00 || [[User:BernardoDalSeno]] || EEG
|-
|2 Feb || 10:00-18:00 || [[User:BernardoDalSeno]] || EEG
|-
|5 Feb || 9:30-18:00 || [[User:TizianoDalbis]] || EEG
|-
|6 Feb || 18:00-24:00 || [[User:BernardoDalSeno]] || EEG
|-
|7 Feb || 0:00-18:00 || [[User:BernardoDalSeno]] || EEG
|-
|19 Feb || 9:30-18:00 || [[User:TizianoDalbis]] || EEG
|-
|26 Feb || 9:30-18:00 || [[User:TizianoDalbis]] || EEG
|-
|}

== Links ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Affective Computing]] page on this Wiki
* [http://www.airlab.elet.polimi.it/index.php/airlab/visitor_info/airlab_iitlab AIRLab - IITLab]

Predictive BCI Speller based on Motor Imagery

2009-01-24T10:49:43Z

TizianoDalbis: /* Part 3: Project tracking */

Predictive BCI Speller based on Motor Imagery

2009-01-05T08:42:21Z

TizianoDalbis: /* Part 3: Project tracking */

Predictive BCI Speller based on Motor Imagery

2009-01-05T08:40:41Z

TizianoDalbis:

IIT-Lab

2008-11-19T16:09:15Z

TizianoDalbis: /* Booking */

== What is the IIT-Lab ==

AIRLab-IITLab is dedicated to activities founded by the Italian Institute of Technology.
The lab hosts activities related to Brain-Computer Interfaces (BCI) and Affective Computing.

=== Location ===
It is located in the Rimembranze di Lambrate building of the Department of Electronics and Information, Via Rimembranze di Lambrate, 14, Milan.

=== Access Rules ===
The access to AIRLab-IITLab is reserved to registered users. If you are student and want to register, you have to fill the AIRLab registration form (to be signed by your tutor) and the security form. The key of the lab is provided to registered users by the doorkeeper at the main entrance of the Lambrate building.

=== Booking ===

Please book the instrument you want to use by adding an entry to the table; the booking of an instrument implies the booking of the room. If you want to use a different instrument at the same time of an existing booking, please contact the other person involved and check that you can share the room; alternatively, you can ask the doorkeeper for an empty room in the building.


{| border="1"
! Day !! Time !! Person !! Instrument
|-
| Thursday 20 November || 10:00-19:30 || [[User:TizianoDalbis]] || EEG
|-
| Thursday 27 November || 14:00-18:30 || [[User:TizianoDalbis]] || EEG
|-
| Thursday 4 December || 14:00-18:30 || [[User:TizianoDalbis]] || EEG
|}

== Links ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Affective Computing]] page on this Wiki
* [http://www.airlab.elet.polimi.it/index.php/airlab/visitor_info/airlab_iitlab AIRLab - IITLab]

IIT-Lab

2008-11-11T17:35:47Z

TizianoDalbis: /* Booking */

== What is the IIT-Lab ==

AIRLab-IITLab is dedicated to activities founded by the Italian Institute of Technology.
The lab hosts activities related to Brain-Computer Interfaces (BCI) and Affective Computing.

=== Location ===
It is located in the Rimembranze di Lambrate building of the Department of Electronics and Information, Via Rimembranze di Lambrate, 14, Milan.

=== Access Rules ===
The access to AIRLab-IITLab is reserved to registered users. If you are student and want to register, you have to fill the AIRLab registration form (to be signed by your tutor) and the security form. The key of the lab is provided to registered users by the doorkeeper at the main entrance of the Lambrate building.

=== Booking ===

Please book the instrument you want to use by adding an entry to the table; the booking of an instrument implies the booking of the room. If you want to use a different instrument at the same time of an existing booking, please contact the other person involved and check that you can share the room; alternatively, you can ask the doorkeeper for an empty room in the building.


{| border="1"
! Day !! Time !! Person !! Instrument
|-
| Wednesday 12 November || 14:00-18:00 || [[User:TizianoDalbis]] || EEG
|-
| Thursday 13 November || 10:00-15:00 || [[User:AndreaSgarlata]] || EEG
|-
| Thursday 13 November || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
|-
| Thursday 20 November || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
|-
| Thursday 27 November || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
|-
| Thursday 4 December || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
|}

== Links ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Affective Computing]] page on this Wiki
* [http://www.airlab.elet.polimi.it/index.php/airlab/visitor_info/airlab_iitlab AIRLab - IITLab]

Predictive BCI Speller based on Motor Imagery

2008-11-09T17:13:04Z

TizianoDalbis: /* '''Part 3: Project tracking''' */

Predictive BCI Speller based on Motor Imagery

2008-11-09T17:11:35Z

TizianoDalbis: /* '''Part 4: Documents''' */

File:Predictive BCI speller v1.pdf

2008-11-09T17:11:07Z

TizianoDalbis: uploaded a new version of "Image:Predictive BCI speller v1.pdf": Description document for the thesis: "Predictive BCI Speller based on motor imagery"

First document with main ideas about Predicrive BCI Speller.

Predictive BCI Speller based on Motor Imagery

2008-11-03T17:53:16Z

TizianoDalbis: /* '''Part 4: Documents''' */

File:Predictive BCI speller v1.pdf

2008-11-03T17:52:44Z

TizianoDalbis: uploaded a new version of "Image:Predictive BCI speller v1.pdf": Description of the thesis: Predictive BCI Speller based on Motor Imagery

First document with main ideas about Predicrive BCI Speller.

Predictive BCI Speller based on Motor Imagery

2008-11-03T17:51:45Z

TizianoDalbis: /* '''Part 3: Project tracking''' */

IIT-Lab

2008-10-31T19:02:00Z

TizianoDalbis: /* Booking */

== What is the IIT-Lab ==

AIRLab-IITLab is dedicated to activities founded by the Italian Institute of Technology.
The lab hosts activities related to Brain-Computer Interfaces (BCI) and Affective Computing.

=== Location ===
It is located in the Rimembranze di Lambrate building of the Department of Electronics and Information, Via Rimembranze di Lambrate, 14, Milan.

=== Access Rules ===
The access to AIRLab-IITLab is reserved to registered users. If you are student and want to register, you have to fill the AIRLab registration form (to be signed by your tutor) and the security form. The key of the lab is provided to registered users by the doorkeeper at the main entrance of the Lambrate building.

=== Booking ===

Please book the instrument you want to use by adding an entry to the table; the booking of an instrument implies the booking of the room. If you want to use a different instrument at the same time of an existing booking, please contact the other person involved and check that you can share the room; alternatively, you can ask the doorkeeper for an empty room in the building.


{| border="1"
! Day !! Time !! Person !! Instrument
|-
| Thursday 6 November || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
|-
| Thursday 13 November || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
|-
| Thursday 20 November || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
|-
|}

== Links ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Affective Computing]] page on this Wiki
* [http://www.airlab.elet.polimi.it/index.php/airlab/visitor_info/airlab_iitlab AIRLab - IITLab]

IIT-Lab

2008-10-31T19:01:32Z

TizianoDalbis: /* Booking */

== What is the IIT-Lab ==

AIRLab-IITLab is dedicated to activities founded by the Italian Institute of Technology.
The lab hosts activities related to Brain-Computer Interfaces (BCI) and Affective Computing.

=== Location ===
It is located in the Rimembranze di Lambrate building of the Department of Electronics and Information, Via Rimembranze di Lambrate, 14, Milan.

=== Access Rules ===
The access to AIRLab-IITLab is reserved to registered users. If you are student and want to register, you have to fill the AIRLab registration form (to be signed by your tutor) and the security form. The key of the lab is provided to registered users by the doorkeeper at the main entrance of the Lambrate building.

=== Booking ===

Please book the instrument you want to use by adding an entry to the table; the booking of an instrument implies the booking of the room. If you want to use a different instrument at the same time of an existing booking, please contact the other person involved and check that you can share the room; alternatively, you can ask the doorkeeper for an empty room in the building.


{| border="1"
! Day !! Time !! Person !! Instrument
|-
| Thursday 6 November || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
-
| Thursday 13 November || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
-
| Thursday 20 November || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
|-
|}

== Links ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Affective Computing]] page on this Wiki
* [http://www.airlab.elet.polimi.it/index.php/airlab/visitor_info/airlab_iitlab AIRLab - IITLab]

IIT-Lab

2008-10-31T19:01:11Z

TizianoDalbis: /* Booking */

== What is the IIT-Lab ==

AIRLab-IITLab is dedicated to activities founded by the Italian Institute of Technology.
The lab hosts activities related to Brain-Computer Interfaces (BCI) and Affective Computing.

=== Location ===
It is located in the Rimembranze di Lambrate building of the Department of Electronics and Information, Via Rimembranze di Lambrate, 14, Milan.

=== Access Rules ===
The access to AIRLab-IITLab is reserved to registered users. If you are student and want to register, you have to fill the AIRLab registration form (to be signed by your tutor) and the security form. The key of the lab is provided to registered users by the doorkeeper at the main entrance of the Lambrate building.

=== Booking ===

Please book the instrument you want to use by adding an entry to the table; the booking of an instrument implies the booking of the room. If you want to use a different instrument at the same time of an existing booking, please contact the other person involved and check that you can share the room; alternatively, you can ask the doorkeeper for an empty room in the building.


{| border="1"
! Day !! Time !! Person !! Instrument
|-
| Thursday 6 November || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
| Thursday 13 November || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
| Thursday 20 November || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
|-
|}

== Links ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Affective Computing]] page on this Wiki
* [http://www.airlab.elet.polimi.it/index.php/airlab/visitor_info/airlab_iitlab AIRLab - IITLab]

Predictive BCI Speller based on Motor Imagery

2008-10-31T18:57:23Z

TizianoDalbis: /* '''Part 4: Documents''' */

Predictive BCI Speller based on Motor Imagery

2008-10-31T18:56:57Z

TizianoDalbis: /* '''Part 3: Project tracking''' */

File:Predictive BCI speller v1.pdf

2008-10-31T18:53:03Z

TizianoDalbis: uploaded a new version of "Image:Predictive BCI speller v1.pdf": Description of the thesis "Predictive BCI Speller based on Motor Imagery"

First document with main ideas about Predicrive BCI Speller.

IIT-Lab

2008-10-27T17:37:38Z

TizianoDalbis: /* Booking */

== What is the IIT-Lab ==

AIRLab-IITLab is dedicated to activities founded by the Italian Institute of Technology.
The lab hosts activities related to Brain-Computer Interfaces (BCI) and Affective Computing.

=== Location ===
It is located in the Rimembranze di Lambrate building of the Department of Electronics and Information, Via Rimembranze di Lambrate, 14, Milan.

=== Access Rules ===
The access to AIRLab-IITLab is reserved to registered users. If you are student and want to register, you have to fill the AIRLab registration form (to be signed by your tutor) and the security form. The key of the lab is provided to registered users by the doorkeeper at the main entrance of the Lambrate building.

=== Booking ===

Please book the instrument you want to use by adding an entry to the table; the booking of an instrument implies the booking of the room. If you want to use a different instrument at the same time of an existing booking, please contact the other person involved and check that you can share the room; alternatively, you can ask the doorkeeper for an empty room in the building.


{| border="1"
! Day !! Time !! Person !! Instrument
|-
|
| Thursday 30 October || 15:00-18:00 || [[User:TizianoDalbis]] || EEG
|-
|}

== Links ==

* [[Brain-Computer Interface]] page on this Wiki
* [[Affective Computing]] page on this Wiki
* [http://www.airlab.elet.polimi.it/index.php/airlab/visitor_info/airlab_iitlab AIRLab - IITLab]