Classification of fNIRS data using wavelets and support vector machine during speed and force imagination

In this paper, we present a method for classifying functional near-infrared spectroscopy (fNIRS) data using wavelets and support vector machine (SVM). fNIRS data is acquired by ETG-4000 during speed and force imagination. Probes location is around C3 and C4 in 10–20 international system. After preprocessing the data using NIRS-SPM, we decompose it with ‘db5’ wavelet for 9 levels to do a multiresolution analysis (MRA). Then, the approximation and detail signal at every level are used for SVM classification using libSVM toolbox. The results show that frequency band between 0.02 and 0.08Hz is important for classification, especially frequency band between 0.02 and 0.04Hz. This finding is useful for building an fNIRS-based brain computer interface (BCI) system.

[1]  Hellmuth Obrig,et al.  Towards a standard analysis for functional near-infrared imaging , 2004, NeuroImage.

[2]  Vladimir Vapnik,et al.  An overview of statistical learning theory , 1999, IEEE Trans. Neural Networks.

[3]  T. Muroga,et al.  Estimation algorithm of tapping movement by NIRS , 2006, 2006 SICE-ICASE International Joint Conference.

[4]  Stéphane Mallat,et al.  A Theory for Multiresolution Signal Decomposition: The Wavelet Representation , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  C. Soraghan,et al.  A 12-Channel, real-time near-infrared spectroscopy instrument for brain-computer interface applications , 2008, 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[6]  Cuntai Guan,et al.  Near infrared spectroscopy based brain-computer interface , 2005, International Conference on Experimental Mechanics.

[7]  G..  A Theory for Multiresolution Signal Decomposition : The Wavelet Representation , 2004 .

[8]  Yong Xu,et al.  Systems and Strategies for Accessing the Information Content of fNIRS Imaging in Support of Noninvasive BCI Applications , 2009, HCI.

[9]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[10]  David A Boas,et al.  Noninvasive measurement of neuronal activity with near-infrared optical imaging , 2004, NeuroImage.

[11]  Bernhard E. Boser,et al.  A training algorithm for optimal margin classifiers , 1992, COLT '92.

[12]  Y. Hoshi,et al.  Spatiotemporal imaging of human brain activity by functional near-infrared spectroscopy , 2001 .

[13]  Sungho Tak,et al.  NIRS-SPM: Statistical parametric mapping for near-infrared spectroscopy , 2009, NeuroImage.

[14]  Sungho Tak,et al.  Wavelet-MDL based detrending method for near infrared spectroscopy (NIRS) , 2008, SPIE BiOS.

[15]  D. Yves von Cramon,et al.  Prefrontal activation due to Stroop interference increases during development—an event-related fNIRS study , 2004, NeuroImage.

[16]  Niels Birbaumer,et al.  Hemodynamic brain-computer interfaces for communication and rehabilitation , 2009, Neural Networks.

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

[19]  C. Markham,et al.  Hemodynamics for Brain-Computer Interfaces , 2008, IEEE Signal Processing Magazine.

[20]  Ron Kohavi,et al.  A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[21]  T. Nakada,et al.  Dorsolateral prefrontal lobe activation declines significantly with age Functional NIRS study , 2003, Journal of Neurology.

[22]  Britton Chance,et al.  Ischemic and bleeding disease monitoring with fNIRS imager: one case report , 1999, Saratov Fall Meeting.

[23]  A. Villringer,et al.  Near infrared spectroscopy (NIRS): A new tool to study hemodynamic changes during activation of brain function in human adults , 1993, Neuroscience Letters.