TDD:Auxiliary framework for recognizing people with depression based on physiological and emotional characteristics of EEG signals

Due to the severe imbalance of the doctor-patient ratio in the world, many patients cannot get mental health diagnoses in time, and there is a lack of objective diagnostic methods for depression at present. To address these issues rapidly and automatically, this paper proposes a new framework, Two-t...

Full description

Saved in:
Bibliographic Details
Published in2022 Euro-Asia Conference on Frontiers of Computer Science and Information Technology (FCSIT) pp. 164 - 169
Main Authors Hu, Bin, Wang, YouMian, Chen, ZhenXiang, Jiang, XiaoQian, Liu, WenJuan, Wang, PeiCheng
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2022
Subjects
Brain modeling
Depression
EEG
Emotion-induced
Feature extraction
Interference
Medical services
Mental health
physical signals
TDD
Watches
Online AccessGet full text

Cover

More Information
Summary:Due to the severe imbalance of the doctor-patient ratio in the world, many patients cannot get mental health diagnoses in time, and there is a lack of objective diagnostic methods for depression at present. To address these issues rapidly and automatically, this paper proposes a new framework, Two-tracks depression diagnosis(TDD), which only requires subjects to watch 60 seconds of emotion-induced material. The framework can extract linear and nonlinear features from Electro-EncephaloGram(EEG) signals for identifying depression. At the same time, it effectively extracts the subjects' emotional features as cues for identifying depression under the guidance of the emotion-induced paradigm. The accuracy of this method in a normal environment can reach 89.03%. The robustness of TDD is obviously better when using cross-individual physiological signals to train the model. Under the condition of using all subjects, the accuracy of TDD is 11.71% higher than that of the model using only physiological signal features. TDD is completely based on objective physiological signals for depression detection, which are difficult to be manipulated subjectively. Therefore, this method can eliminate subjective interference and make the results more realistic and objective than scale detection.
DOI:10.1109/FCSIT57414.2022.00041