Limits...
Towards zero training for brain-computer interfacing.

Krauledat M, Tangermann M, Blankertz B, Müller KR - PLoS ONE (2008)

Bottom Line: This way the calibration periods can be dramatically shortened or even completely omitted for these 'experienced' BCI users.The feasibility of our novel approach is demonstrated with a series of online BCI experiments.Although performed without any calibration measurement at all, no loss of classification performance was observed.

View Article: PubMed Central - PubMed

Affiliation: Machine Learning Laboratory, Berlin Institute of Technology, Berlin, Germany. kraulem@cs.tu-berlin.de

ABSTRACT
Electroencephalogram (EEG) signals are highly subject-specific and vary considerably even between recording sessions of the same user within the same experimental paradigm. This challenges a stable operation of Brain-Computer Interface (BCI) systems. The classical approach is to train users by neurofeedback to produce fixed stereotypical patterns of brain activity. In the machine learning approach, a widely adapted method for dealing with those variances is to record a so called calibration measurement on the beginning of each session in order to optimize spatial filters and classifiers specifically for each subject and each day. This adaptation of the system to the individual brain signature of each user relieves from the need of extensive user training. In this paper we suggest a new method that overcomes the requirement of these time-consuming calibration recordings for long-term BCI users. The method takes advantage of knowledge collected in previous sessions: By a novel technique, prototypical spatial filters are determined which have better generalization properties compared to single-session filters. In particular, they can be used in follow-up sessions without the need to recalibrate the system. This way the calibration periods can be dramatically shortened or even completely omitted for these 'experienced' BCI users. The feasibility of our novel approach is demonstrated with a series of online BCI experiments. Although performed without any calibration measurement at all, no loss of classification performance was observed.

Show MeSH
The feedback results for each of the six subjects.The feedback accuracy is denoted for the 100 trials of each run. The initial three runs, here marked as block “I”, were done with the Zero-Training classifier, and in the following the order of the classifiers was randomly permuted in each block of two runs, here denoted as “II–V”. The shift of the blue curve relative to the green curve within the shaded areas indicates the order of the classifiers within the block.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2500157&req=5

pone-0002967-g004: The feedback results for each of the six subjects.The feedback accuracy is denoted for the 100 trials of each run. The initial three runs, here marked as block “I”, were done with the Zero-Training classifier, and in the following the order of the classifiers was randomly permuted in each block of two runs, here denoted as “II–V”. The shift of the blue curve relative to the green curve within the shaded areas indicates the order of the classifiers within the block.

Mentions: The first three runs of feedback showed that all subjects under study were able to operate the BCI with the pre-computed classifier at a high accuracy (only 10 trials per class from the current day were required to update the bias for the classification scenario). For every subject Fig. 4 shows the percentage of successful (“hit”) trials from each run. After the third run, the subjects could not know in advance, which one of the two classifiers (Zero-Training or ordinary CSP) was used for the generation of the feedback.


Towards zero training for brain-computer interfacing.

Krauledat M, Tangermann M, Blankertz B, Müller KR - PLoS ONE (2008)

The feedback results for each of the six subjects.The feedback accuracy is denoted for the 100 trials of each run. The initial three runs, here marked as block “I”, were done with the Zero-Training classifier, and in the following the order of the classifiers was randomly permuted in each block of two runs, here denoted as “II–V”. The shift of the blue curve relative to the green curve within the shaded areas indicates the order of the classifiers within the block.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2500157&req=5

pone-0002967-g004: The feedback results for each of the six subjects.The feedback accuracy is denoted for the 100 trials of each run. The initial three runs, here marked as block “I”, were done with the Zero-Training classifier, and in the following the order of the classifiers was randomly permuted in each block of two runs, here denoted as “II–V”. The shift of the blue curve relative to the green curve within the shaded areas indicates the order of the classifiers within the block.
Mentions: The first three runs of feedback showed that all subjects under study were able to operate the BCI with the pre-computed classifier at a high accuracy (only 10 trials per class from the current day were required to update the bias for the classification scenario). For every subject Fig. 4 shows the percentage of successful (“hit”) trials from each run. After the third run, the subjects could not know in advance, which one of the two classifiers (Zero-Training or ordinary CSP) was used for the generation of the feedback.

Bottom Line: This way the calibration periods can be dramatically shortened or even completely omitted for these 'experienced' BCI users.The feasibility of our novel approach is demonstrated with a series of online BCI experiments.Although performed without any calibration measurement at all, no loss of classification performance was observed.

View Article: PubMed Central - PubMed

Affiliation: Machine Learning Laboratory, Berlin Institute of Technology, Berlin, Germany. kraulem@cs.tu-berlin.de

ABSTRACT
Electroencephalogram (EEG) signals are highly subject-specific and vary considerably even between recording sessions of the same user within the same experimental paradigm. This challenges a stable operation of Brain-Computer Interface (BCI) systems. The classical approach is to train users by neurofeedback to produce fixed stereotypical patterns of brain activity. In the machine learning approach, a widely adapted method for dealing with those variances is to record a so called calibration measurement on the beginning of each session in order to optimize spatial filters and classifiers specifically for each subject and each day. This adaptation of the system to the individual brain signature of each user relieves from the need of extensive user training. In this paper we suggest a new method that overcomes the requirement of these time-consuming calibration recordings for long-term BCI users. The method takes advantage of knowledge collected in previous sessions: By a novel technique, prototypical spatial filters are determined which have better generalization properties compared to single-session filters. In particular, they can be used in follow-up sessions without the need to recalibrate the system. This way the calibration periods can be dramatically shortened or even completely omitted for these 'experienced' BCI users. The feasibility of our novel approach is demonstrated with a series of online BCI experiments. Although performed without any calibration measurement at all, no loss of classification performance was observed.

Show MeSH