Riemannian Geometry on Connectivity for Clinical BCI

Riemannian BCI based on EEG covariance have won many data competitions and achieved very high classification results on BCI datasets. To increase the accuracy of BCI systems, we propose an approach grounded on Riemannian geometry that extends this framework to functional connectivity measures. This paper describes the approach submitted to the Clinical BCI Challenge-WCCI2020 and that ranked 1st on the task 1 of the competition.

[1]  Kon Max Wong,et al.  Electroencephalogram signals classification for sleep-state decision - a Riemannian geometry approach , 2012, IET Signal Process..

[2]  Klaus-Robert Müller,et al.  Neurophysiological predictor of SMR-based BCI performance , 2010, NeuroImage.

[3]  David H. Wolpert,et al.  Stacked generalization , 1992, Neural Networks.

[4]  Alexandre Barachant,et al.  Riemannian geometry for EEG-based brain-computer interfaces; a primer and a review , 2017 .

[5]  An H. Do,et al.  Operation of a brain-computer interface walking simulator for individuals with spinal cord injury , 2013, Journal of NeuroEngineering and Rehabilitation.

[6]  Richard M. Leahy,et al.  Brainstorm: A User-Friendly Application for MEG/EEG Analysis , 2011, Comput. Intell. Neurosci..

[7]  Danielle S. Bassett,et al.  Phase/Amplitude Synchronization of Brain Signals During Motor Imagery BCI Tasks , 2019, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[8]  Dennis J. McFarland,et al.  Brain–computer interfaces for communication and control , 2002, Clinical Neurophysiology.

[9]  M. Molinari,et al.  Brain–computer interface boosts motor imagery practice during stroke recovery , 2015, Annals of neurology.

[10]  Sugiyama Masashi,et al.  Importance-Weighted Covariance Estimation for Robust Common Spatial Pattern , 2014 .

[11]  F. Yger,et al.  Riemannian Approaches in Brain-Computer Interfaces: A Review , 2017, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[12]  N. Higham.  COMPUTING A NEAREST SYMMETRIC POSITIVE SEMIDEFINITE MATRIX , 1988 .

[13]  Christian Jutten,et al.  The Riemannian Minimum Distance to means field Classifier , 2019, GBCIC.

[14]  Danielle S. Bassett,et al.  Functional disconnection of associative cortical areas predicts performance during BCI training , 2020, NeuroImage.

[15]  Masashi Sugiyama,et al.  Supervised LogEuclidean Metric Learning for Symmetric Positive Definite Matrices , 2015, ArXiv.

[16]  M Congedo,et al.  A review of classification algorithms for EEG-based brain–computer interfaces: a 10 year update , 2018, Journal of neural engineering.

[17]  Sung Chan Jun,et al.  High Theta and Low Alpha Powers May Be Indicative of BCI-Illiteracy in Motor Imagery , 2013, PloS one.

[18]  Masashi Sugiyama,et al.  Geometry-aware principal component analysis for symmetric positive definite matrices , 2017, Machine Learning.

[19]  M. Corbetta,et al.  Large-scale cortical correlation structure of spontaneous oscillatory activity , 2012, Nature Neuroscience.

[20]  Margaret C. Thompson,et al.  Critiquing the Concept of BCI Illiteracy , 2018, Science and Engineering Ethics.

[21]  Christian Jutten,et al.  Multiclass Brain–Computer Interface Classification by Riemannian Geometry , 2012, IEEE Transactions on Biomedical Engineering.

[22]  W. Klimesch.  EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis , 1999, Brain Research Reviews.

[23]  Jonas Richiardi,et al.  Graph analysis of functional brain networks: practical issues in translational neuroscience , 2014, Philosophical Transactions of the Royal Society B: Biological Sciences.

[24]  F. Varela,et al.  Measuring phase synchrony in brain signals , 1999, Human brain mapping.

[25]  Christian Jutten,et al.  Riemannian Geometry Applied to BCI Classification , 2010, LVA/ICA.