2014 Volume 7 Issue 6 Pages 327-331
The goal of this research is to develop a rehabilitation system based on a brain-machine interface (BMI) for paraplegic patients. This requires the ability to detect a patient's desire on the basis of his or her brain activity. In the work reported here, the authors used near-infrared spectroscopy (NIRS) to measure the brain activity of healthy subjects while they exercised on a bicycle ergometer. Analysis results showed that the oxy-Hb level increased at the start of the exercise. This variation in the oxy-Hb level occurred a few seconds after the start of exercise. On the basis of the analysis results, the authors considered a pedaling exercise state discrimination that uses the oxy-Hb level and its time derivative. The discrimination results for three healthy subjects showed over 72% accuracy. For a paraplegic patient's pedal exercise imagery, this discrimination scheme still had 74% accuracy.