The stroke rehabilitation movement identification method that incorporates convolutional neural networks and long and short-term memory neural networks is proposed to address low recognition rate,poor perceived ease of use of intelligent self-rehabilitation behaviors in stroke patients.Firstly,the convolutional neural network model is used to obtain feature information about the data local space,and the long and short-term memory neural network model is further used to obtain feature information about the data local long-term correlation.The two fuse sensor data from the sensor data of the smart mobile terminal to obtain precise information on the temporal and spatial the characteristics related to limb changes in rehabilitation training.Finally,the behavioral classification of stroke rehabilitation is achieved.Experiment analysis shows that the proposed method has higher accuracy than traditional methods in identifying patient behav-iors.It has some application value in rehabilitation training.
关键词
移动终端/卷积神经网络/长短期记忆网络/脑卒中康复/行为识别
Key words
mobile terminal/Convolutional Neural Network/Short and Long-term Memory Network/stroke rehabilitation/behavioral identification
维普
万方数据