A Novel Fuzzy Neural Network-based Rehabilitation Stage Classifying Method for the Upper Limb Rehabilitation Robotic System

摘要

According to the “brain plasticity” theory, the hemiplegia patients could improve the ability to keep balance through rehabilitation training. This paper proposes a method of collecting sEMG signals and interaction force signals during the rehabilitation training which is completed by continuously making specified actions driven by external forces. The signals are collected to classify the rehabilitation stage and evaluate the rehabilitation effect to solve the problem of stroke rehabilitation physician's shortage. This paper conducted an experiment of rehabilitation exercise based on the manual muscle testing (MMT) scale, fitted a fuzzy neural network and verified its predicted performance for the assessment of rehabilitation stage.