fNIRS-based classification of hand-related motor activity and motor imagery
Alexander Hramov, Innopolis University
Vadim Grubov, Innopolis University
Badarin Artem, innopolis University
Abstract
In the present study we analyzed sensor-level human brain activity during hand-related motor activity and motor imagery using functional near-infrared spectroscopy (fNIRS). We studied blood oxygenation and deoxygenation spatial dynamics with its pronounced hemispheric lateralization when performing motor tasks with left and right hands. This fact allowed us to reveal biomarkers of hemodynamical response in the motor cortex, and use them for classification of the type of movement. The recognition accuracy of real movements is close to 100%, while the classification accuracy of imaginary movements is lower but still quite high (close to 90%). The advantage of the proposed method is its ability to reliably classify motor activity and motor imagery with generalized parameters, appropriate for different subjects. Possible application for the proposed system lies in the field of neurorehabilitation after severe brain injuries, including traumas and strokes.
Speaker
Vadim Grubov
Innopolis University
Russia
Discussion
Ask question
