An automatic analysis method of sleep structure based on human muscle surface electrical signals

Abstract

The sleep structure is an important indicator reflecting sleep quality. The invention discloses an automatic sleep structure analysis method based on human surface electromyographic signals to solve the problems of expensive monitoring equipment, operation complexity, analysis process complexity, higher technician requirements and the like of a currently common evaluation standard based on polysomnography (PSG). According to the method, firstly, acquired original surface electromyographic signals are fragmented and processed to be divided into active signals and inactive signals, the time domain feature of each frame of inactive electromyographic signal and the frequency domain feature of each inactive electromyographic signal are acquired, on the basis, a classification module is trainedand tested by a random forest method, active fragments are divided into short-time, medium-time and long-time active fragments according to relative amplitude and duration of the active fragments, theelectromyographic signals are classified by the random forest method and the type of the active fragment, and a primary classification result is corrected.

Description

technical field [0001] The invention relates to the field of sleep structure analysis, in particular to a method for automatically analyzing sleep structure by using human muscle surface electrical signals. Background technique [0002] Sleep plays an important role in the physiological function and psychological state of the human body. Poor sleep quality will not only affect the physiological functions of the important organs and systems of the human body, but also cause hypertension, coronary heart disease, diabetes, stroke, depression, anxiety and other diseases, causing many diseases. Damage to system organ function is also an important cause of memory loss and reaction ability reduction, and it can also have a negative impact on operating ability and traffic safety. It is an easily overlooked factor affecting social and economic development. [0003] Sleep structure is an important indicator of sleep quality. During the whole night's sleep, it usually includes 4-6 sle...

AI Technical Summary

Problems solved by technology

Although EEG signals can more accurately reflect the activities of different brain regions, because the signals are weak and susceptible to interference, the monitoring process has high requirements for equipment, monitoring technology and environment, and it is difficult to widely popularize in grassroots medical institutions.

Images