Identification and analysis of limb rehabilitation signal based on wavelet transform

The development of science and technology has promoted the extensive application of surface electromyography (sEMG) collection technique in real-time exercise testing, assistive judgment of rehabilitation therapy, and assessment of intelligent artificial limb application. However, there is a severe lacking of studies on pattern recognition based on effective signal, and evaluation of limb rehabilitation status. To make up for the gap, this paper explores the identification and analysis of limb rehabilitation signal based on wavelet transform. Specifically, the authors detailed the basic flow of sEMG signal generation in motor unit during limb rehabilitation exercise, and proposed a limb EMG pattern recognition method. Then, support vector machine (SVM) was selected to recognize the pattern of the EMG signal extracted from the limb rehabilitation exercise of patients, and to judge the rehabilitation status. Finally, wavelet thresholding was combined with total variation denoising (TVD) to effectively remove the noise from EMG signal. The proposed method was proved effective through experiments.