A Preprocessing Method for EMG Based on Empirical Mode Decomposition

Abstract

The invention discloses an electromyogram preprocessing method based on empirical mode decomposition. The electromyogram preprocessing method includes the steps (1), smoothing electromyogram signals by linear-phase limited long-pulse response filters with different cut-off frequencies; (2), subjecting each smoothed signal to EMD to obtain an IMF corresponding to each signal; (3), taking entropies and extreme values of all the acquired IMF as characteristic parameters for electromyogram analysis. The electromyogram preprocessing method based on the empirical mode decomposition has the advantages that the quantity and the type of the characteristic parameters are decided by the quantity of the filters and the difference among the cut-off frequencies, and accordingly, flexibility is achieved; the more the characteristic parameters are, the more electromyogram information are acquired; in later application, the relation between the electromyogram characteristic parameters and to-be-detected nerves or muscles or other functions is created, and reference for medical diagnosis, daily body examination and the like is provided.

Description

technical field [0001] The invention relates to electromyogram processing technology, more specifically, relates to a preprocessing method for electromyogram based on empirical mode decomposition. Background technique [0002] Electromyography (EMG for short) is the application of electronic instruments to record the electrical activity of muscles at rest or contraction, and the application of electrical stimulation to examine the excitation and conduction function of nerves and muscles. Electromyography can be used to record nerve and muscle activity to determine its function. There will be tiny bioelectrical changes during muscle movement, and electromyography is used as a non-invasive detection method to detect such bioelectrical changes. The amplitude, frequency and other indicators of EMG will change accordingly during exercise, and these changes are closely related to exercise mode, exercise state, and muscle fatigue. It can be seen that the application of electromyo...

AI Technical Summary

Problems solved by technology

However, the parameters selected by the current EMD decomposition for EMG processing are fixed, and there is still a lot of noise mixed in, the features are not obvious enough, and the number of selected features is not flexible enough.

Images