摘要
精神障碍疾病临床上的诊断缺乏客观性的参数,诊断和治疗都面临挑战.随着脑电图机在医院中的普及,脑电数据被广泛应用于精神障碍疾病的潜在生物标志物的发掘.相比传统的群体水平显著性差异分析,深度学习模型有利于实现个体化和智能化预测.文章将综述基于深度学习的精神障碍疾病的脑电信号的研究进展,从深度学习算法到精神障碍患者脑电图的自动分类等方面进行总结和分析.目前,已有研究还存在样本量小等局限,未来可通过数据采集、数据增强等方面进行改进,以期获得更鲁棒的结果,能够运用于临床诊疗中.
The neuropathological clinical diagnosis lacks objective parameters,and both diagnosis and treatment face challenges.With the popularity of electroencephalogram machines in hospitals,EEG data are widely used to uncover potential biomarkers of mental disorders.Compared with the traditional group-level significant difference analysis,deep learning models facilitate individualized and intlligent prediction.The article will review the progress of deep learning-based EEG signals researches in mental disorder diseases,from deep learning algorithms to automatic classification of EEG in mental disorder patients,etc.The article will summarize and analyze the research progress.Currently,the studies still have limitations such as small sample size.Future improvements can be made through data collection and data enhancement to obtain more robust results that can be used in clinical diagnosis and treatment.
作者
赵齐岳
朱耿
过晓洋
杜晓微
王艳
李晓欧
Zhao Qiyue;Zhu Geng;Guo Xiaoyang;Du Xiaowei;Wang Yan;Li Xiaoou(School of Medical Institute,Shanghai University of Medicine&Health Science,Shanghai 201318,China)
出处
《现代仪器与医疗》
CAS
2022年第5期82-85,共4页
Modern Instruments & Medical Treatment
基金
上海健康医学院大学生创新创练计划项目资助(S202210262145)。
作者简介
第一作者:赵齐岳,男,在读本科生,研究方向:生物医学工程,E-mail:960620441@qq.com;通讯作者:李晓欧,男,博士,教授,研究方向:生物医学工程,E-mail:lixo@sumhs.edu.cn。
