Prediction method of lower extremity joint angle based on EMG wavelet correlation dimension

Abstract

The invention relates to a method for predicting joint angles of lower limbs based on myoelectric wavelet correlation dimension. First, the surface electromyography signal is collected from the relevant muscle groups of the lower limbs of the human body, and the action signal segment of the surface electromyography signal is determined by using the energy threshold. Perform wavelet noise reduction on the surface electromyography signal of the action signal segment to obtain an effective surface electromyography signal. Then, the effective surface electromyography signal is decomposed by wavelet multi-scale, and the low-frequency coefficients of each layer are extracted, and then the correlation dimension is calculated for each layer of low-frequency coefficients. Combining the low-frequency coefficient and the correlation dimension, the wavelet correlation dimension coefficient feature of the effective EMG signal is calculated, and this feature is used as the input of the prediction network. Firstly, the extracted EMG signals are divided into training set and test set, and the features are extracted according to the above method. After training the network with the training set, use the test set to verify the prediction accuracy. Experimental results show that this method has a higher prediction rate of knee joint angle in human lower limb movement, and the prediction result is better than other prediction methods.

Description

technical field [0001] The invention belongs to the field of pattern recognition, and relates to a pattern recognition method based on electromyographic signals, in particular to a method for predicting joint angles of human lower limbs based on wavelet correlation dimension features of electromyographic signals. Background technique [0002] Patients with spinal cord injury (spinal cord injury SCI) refer to those who have lost their motor function due to nerve damage, and their postoperative rehabilitation treatment has a long way to go. Rehabilitation training using passive methods such as treadmills and knee extension and flexion is a traditional treatment method, but the therapeutic effect of this method is limited. Practice has proved that active training can improve the reorganization of the cerebral cortex, which is conducive to the recovery of neurons in patients. The traditional human-computer interaction system based on program control restricts the development of...

AI Technical Summary

Problems solved by technology

However, the model has a complex structure and contains many physiological parameters that cannot be directly measured

Images