An Intelligent Biofeedback System Based on Pattern Recognition and Electroacupuncture Imitating Traditional Chinese Medical Acupuncture

This paper presented a novel intelligent biofeedback system based on the pattern recognition of physiological signals and electroacupuncture that imitates the traditional Chinese medical acupuncture. The system is composed of circuit of signal acquisition, computer, intelligent electroacupuncture and relevant softwares. The main purpose of research is to make decisions and judgments by artificial intelligence technique instead of physicians, that is, by the means of feature extraction and pattern recognition of GSR signal and EMG signal, the system can make diagnoses and judgments automatically, and then control the output of intelligent electroacupuncture which is used to speed up the establishing of biofeedback and treat AD (anxiety disorder). In this paper, the ICA (independent component analysis) and RBF (radial basis function) neural network methods were adopted for signal feature extraction and pattern recognition. Although conceived mainly for GSR and EMG biofeedback, and for the rehabilitation of AD patients, the system is flexible to be used with some other physiological variables and to cure other diseases