A CNN-LSTM Hybrid Model for Wrist Kinematics Estimation Using Surface Electromyography

Convolutional neural network (CNN) has been widely exploited for simultaneous and proportional myoelectric control due to its capability of deriving informative, representative, transferable features from surface electromyography (sEMG). However, muscle contractions have strong temporal dependencies, but conventional CNN can only exploit spatial correlations. Considering that the long short-term memory (LSTM) is able capture long-term nonlinear dynamics time-series data, in this article, we propose a CNN-LSTM hybrid model fully explore temporal-spatial information sEMG. First, utilized extract deep sEMG spectrum, then, these are processed via LSTM-based sequence regression estimate wrist kinematics. Six healthy participants recruited participatory collection motion analysis under various experimental setups. Estimation results both intrasession intersession evaluations illustrate significantly outperforms CNN, LSTM, several representative machine learning approaches, particularly when complex movements activated.