Depression recognition and analysis system based on resting state brain network

Abstract

The invention provides a depression recognition and analysis system based on a resting state brain network. The depression recognition and analysis system comprises (a) a resting state electroencephalogram data acquisition and preprocessing module used for collecting resting state electroencephalogram data of subjects and preprocessing the collected resting state electroencephalogram data, (b) a brain network metric extraction module used for constructing a personalized brain network structure, respectively finding out common active brain regions of a depression group and a normal control group from the personalized brain network structure, finding out different brain regions based on the common active brain regions of the two groups and extracting brain network metrics, and (c) a classification recognition module used for conducting feature selection on the extracted brain network metrics and functional connection features and classifying the data of which features are already screened to achieve recognition of depression patients and normal subjects. The depression recognition and analysis system has the advantages that feature dimension is effectively reduced, calculation efficiency is improved, and depression recognition can be effectively achieved.

Description

technical field [0001] The invention relates to the field of network analysis and medical assistance research, in particular to a depression identification and analysis system based on a resting state brain network. Background technique [0002] Depression is a common mental illness characterized clinically by marked and persistent low mood, manifested by lack of interest in life, insomnia or hypersomnia, lack of energy, inability to concentrate, feelings of worthlessness, and feelings of guilt and ruminations about suicide. Depression currently affects more than 350 million people worldwide. The results of the World Mental Health Survey in 17 countries showed that on average about 1 in 20 people reported having a depressive episode in the previous year. According to World Health Organization estimates, by 2020 depression will become the second leading disease in the world. Therefore, timely detection of depression and understanding of the neural mechanisms of depression ...

AI Technical Summary

Problems solved by technology

However, there is no gold standard for the detection of depression today, and a combination of doctor’s consultation and scale is still used. The existing problems are: low patient cooperation, strong subjective bias, low sensitivity and low accuracy

However, the brain has individual differences. Currently, the extraction of network metrics for classification is based on the entire brain network of each subject, which brings problems of high feature dimension and large amount of calculation.

Images