Comparison of Classification Accuracies Between Different Brain Areas During a Two-Class Motor Imagery in a fNIRS Based BCI**This work was supported by a Discovery Grant from the Natural Science and Engineering Research Council of Canada [NSERC RGPIN 2016-04669]

This study examined functional near infrared spectroscopy (fNIRS) data from 29 participants to determine whether the classification accuracy associated with a motor imagery task depended on different areas of the brain. The fNIRS was used to measure concentration changes of oxy- (HbO) and deoxyhemoglobin (HbR). The averages of HbO (mHbO) and HbR (mHbR) were used as features, and linear discriminant analysis (LDA) and support vector machine (SVM) were used as classifiers. The results showed significantly higher classification accuracies for the motor cortex than both frontal and occipital areas, but less accuracy compared to all channels. Furthermore, while SVM resulted in higher accuracy than LDA, mHbO and mHbR led to similar accuracies.

[1]  Filippo Zappasodi,et al.  Deep learning for hybrid EEG-fNIRS brain–computer interface: application to motor imagery classification , 2018, Journal of neural engineering.

[2]  Dean J Krusienski,et al.  Brain-computer interfaces in medicine. , 2012, Mayo Clinic proceedings.

[3]  Eda Akman Aydin,et al.  Subject-Specific feature selection for near infrared spectroscopy based brain-computer interfaces , 2020, Comput. Methods Programs Biomed..

[4]  Amir H. Moslehi,et al.  Discrimination of Two-Class Motor Imagery in a fNIRS Based Brain Computer Interface , 2020, 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC).

[5]  Cuntai Guan,et al.  Real-Time Subject-Independent Pattern Classification of Overt and Covert Movements from fNIRS Signals , 2016, PloS one.

[6]  Grzegorz M. Wójcik,et al.  Most Popular Signal Processing Methods in Motor-Imagery BCI: A Review and Meta-Analysis , 2018, Front. Neuroinform..

[7]  F. Jöbsis.  Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. , 1977, Science.

[8]  Keum-Shik Hong,et al.  fNIRS-based brain-computer interfaces: a review , 2015, Front. Hum. Neurosci..

[9]  Hong Peng,et al.  Single-Trial Classification of fNIRS Signals in Four Directions Motor Imagery Tasks Measured From Prefrontal Cortex , 2018, IEEE Transactions on NanoBioscience.

[10]  Yingchun Zhang,et al.  Enhancing Performance of a Hybrid EEG-fNIRS System Using Channel Selection and Early Temporal Features , 2017, Front. Hum. Neurosci..

[11]  D. Giavarina.  Understanding Bland Altman analysis , 2015, Biochemia medica.

[12]  M. Erb,et al.  Activation of Cortical and Cerebellar Motor Areas during Executed and Imagined Hand Movements: An fMRI Study , 1999, Journal of Cognitive Neuroscience.

[13]  Klaus-Robert Müller,et al.  Open Access Dataset for EEG+NIRS Single-Trial Classification , 2017, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[14]  Gernot R. Müller-Putz,et al.  Imagine squeezing a cactus: Cortical activation during affective motor imagery measured by functional near-infrared spectroscopy , 2018, Brain and Cognition.

[15]  Shirley M Coyle,et al.  Brain–computer interface using a simplified functional near-infrared spectroscopy system , 2007, Journal of neural engineering.

[16]  S. Arridge,et al.  Estimation of optical pathlength through tissue from direct time of flight measurement , 1988 .

[17]  Cuntai Guan,et al.  Temporal classification of multichannel near-infrared spectroscopy signals of motor imagery for developing a brain–computer interface , 2007, NeuroImage.

[18]  Guideline Thirteen: Guidelines for Standard Electrode Position Nomenclature , 1994, Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society.

[19]  Hasan Onur Keles,et al.  Hybrid EEG-fNIRS Asynchronous Brain-Computer Interface for Multiple Motor Tasks , 2016, PloS one.

[20]  Ganesan Venkatasubramanian,et al.  Investigation of deep convolutional neural network for classification of motor imagery fNIRS signals for BCI applications , 2020, Biomed. Signal Process. Control..

[21]  Martin Wolf,et al.  Single-trial classification of motor imagery differing in task complexity: a functional near-infrared spectroscopy study , 2011, Journal of NeuroEngineering and Rehabilitation.

[22]  K. Mankodiya,et al.  Hand Motion Detection in fNIRS Neuroimaging Data , 2017, Healthcare.

[23]  D. Markland,et al.  Movement imagery ability: development and assessment of a revised version of the vividness of movement imagery questionnaire. , 2008, Journal of sport & exercise psychology.

[24]  Shirley Coyle,et al.  On the suitability of near-infrared (NIR) systems for next-generation brain-computer interfaces. , 2004, Physiological measurement.

[25]  G. Flandin,et al.  Sensor space group analysis for fNIRS data , 2016, Journal of Neuroscience Methods.

[26]  J. Leon-Carrion,et al.  Functional Near-Infrared Spectroscopy (fNIRS): Principles and Neuroscientific Applications , 2012 .

[27]  Qing Yang,et al.  A Brain-Computer Interface Based on a Few-Channel EEG-fNIRS Bimodal System , 2017, IEEE Access.

[28]  D. Altman,et al.  Measuring agreement in method comparison studies , 1999, Statistical methods in medical research.

[29]  Gang Shi,et al.  Classification of Hemodynamic Responses Associated With Force and Speed Imagery for a Brain-Computer Interface , 2015, Journal of Medical Systems.