Sleep stage classification method based on complex networks and deep learning and application thereof

Abstract

A sleep stage classification method based on complex networks and deep learning and application thereof are provided. The sleep stage classification method comprises: acquiring characteristic indexesof a finite passage visibility graph complex network, to be specific, establishing the finite passage visibility graph complex network to extract node degrees, acquiring a node degree sequence according to the node degrees, and applying the node degree sequence as a characteristic index of the finite passage visibility graph complex network; establishing finite passage visibility graph complex networks having a finite passage visual range of 1 for sleep stage electroencephalography fragments; classifying the sleep stage electroencephalogram fragments into four classes according to awake stage,light sleep stage, deep sleep stage and rapid eye movement stage by means of 10-fold cross validation and a long short term memory model. The method of the invention is applicable to head-set intelligent wearable devices; by analyzing sleep electroencephalogram measured by the intelligent wearable devices, it is possible to understand the brain state of a user and provide necessary early warnings.

Description

technical field [0001] The invention relates to a sleep stage monitoring method and application. In particular, it relates to a sleep stage classification method and application based on complex network and deep learning. Background technique [0002] Sleep is an important, dynamic and regular process that occurs in human beings. It has an important impact on a person's daily activities. Sleep is one of the most important functions of brain activity. A healthy human brain goes through several psycho-physiological states, known as sleep stages, during sleep. With the increasing pressure of social life, more and more people are troubled by sleep-related diseases, such as sleep apnea, insomnia, lethargy, and fainting. This will seriously affect people's health and quality of life. More and more people hope to understand their sleep state through continuous EEG signal collection and analysis and sleep monitoring. Sleep stage detection is to classify sleep stages, which plays ...

AI Technical Summary

Problems solved by technology

However, most of the training models in the past were shallow, and the classification effect could not be very accurate.

Images