A machine learning classification study of patients with anxiety disorders based on EEG characteristics
Journal Title: Journal of Air Force Medical University - Year 2023, Vol 44, Issue 10
Abstract
Objective To explore the electroencephalogram (EEG) characteristics of patients with anxiety disorders when answering questionnaires and opening and closing their eyes, so as to provide technical support for military psychological selection and multimodal fusion theory. Methods A total of 54 subjects were collected and divided into anxiety disorders group (24 subjects) and normal group (30 subjects). It was found that power spectral density ( PSD) could be used to evaluate brain abnormalities in patients with anxiety disorders in frequency domain analysis of the two groups. Results ① The amplitude of EEG PSD of the two groups was significantly different in the low-frequency band, and the anxiety disorders group was higher than the normal group. ② The full frequency band was significant in the eye-opening state, and the inhibitory effect of eye-opening appeared in the alpha band. ③ By classifying the population through machine learning, the recognition rate of multimodal fusion index increased by 5% compared with the single behavioral index. Conclusion The high-risk group of the anxiety disorders group does not meet the criteria for a diagnosis of anxiety disorders, but it is easy to induce anxiety disorders clinically. The feature extraction and population classification of EEG frequency domain indicators by using machine learning can improve the recognition of people with anxiety disorders, which has forward-looking significance in future personnel selection and clinical evaluation.
Authors and Affiliations
FENG Tingwei, REN Lei, WU Lin, LI Danyang, YANG Wei, ZHANG Peng, WANG Buyao, WANG Hui, LIU Xufeng
Keywords
Related Articles
- EP ID EP723831
- DOI -
- Views 53
- Downloads 1
Analysis of cardiac CT imaging and clinical characteristics of false aneurysm of the membranous septum with ventricular septal defect
Objective To analyze the cardiac CT angiography ( CTA ) imaging and clinical characteristics in patients with false aneurysm of the membranous septum ( FAMS) with ventricular septal defect ( VSD). Methods The cardiac C...
Cannabinoid receptor agonist ACEA promotes myelin recovery after spinal cord injury by regulating IL-33
Objective To investigate the protective effect of activating cannabinoid type 1 receptor (CB1R) agonist ACEA in promoting myelin recovery after spinal cord injury (SCI) in mice, and to clarify the role of interleukin-33...
Design and development of “ Man-Machine System ” training system based on scaffolding teaching method
Objective To design and develop a training system for classroom teaching based on the advanced smart classroom hardware platform and scaffolding teaching method of Air Force Medical University, aiming at the teaching co...
Visualization analysis of SGLT2 inhibitors in the treatment of heart failure based on CiteSpace
Objective To visually analyze the number of papers, authors, research institutions and keywords related to sodium-glucose cotransporter 2 (SGLT2) inhibitors in the treatment of heart failure ( HF) by using CiteSpace, an...
Identification and validation of glycolysis-related biomarkers in ovarian cancer based on bioinformatics and machine learning
Objective To explore potential biomarkers and therapeutic targets with diagnostic value for ovarian cancer (OC) from genes related to glycolysis signaling pathways based on bioinformatics and machine learning. Methods...