基于多模态情感脑机接口的抑郁症客观评估与调控治疗  

Objective assessment and treatment of depression based on multimodal affective brain computer interfaces

作　　者：吕宝粮[1] 张亚倩 刘伟[1] 郑伟龙 Lu Bao-Liang;Zhang Yaqian;Liu Wei;Zheng Wei-Long(The Center for Brain-Like Computing and Machine Intelligence,Department of Computer Science and Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;The Department of Brain and Cognitive Science,Massachusetts Institute of Technology,Cambridge,MA 02139,USA)

机构地区：[1]上海交通大学计算机科学与工程系仿脑计算与机器智能研究中心,200240 [2]麻省理工学院脑与认知科学系,剑桥02139

出　　处：《中华精神科杂志》2021年第4期243-251,共9页Chinese Journal of Psychiatry

基　　金：国家重点研发计划(2017YFB1002500);国家自然科学基金面上项目(61673266)。

摘　　要：本文探讨多模态情感脑机接口在抑郁症客观评估和难治性抑郁症脑深部电刺激治疗中的应用。在抑郁症客观评估方面,首先将传统的量表转化成情感交互试验。基于多模态情感脑机接口,同步采集脑电和眼动等多模态数据。通过多模态深度学习和迁移学习等深度学习技术,构建可精确区分抑郁状态的客观评估系统。在难治性抑郁症脑深部电刺激神经调控治疗方面,基于多模态情感脑机接口和强化学习算法,实现脑深部电刺激刺激参数的自动调节和个性化,提升难治性抑郁症治疗的效果。This paper explores the application of multimodal affective brain-computer interfaces(aBCI)in the diagnosis based on the objective assessment of depression and treatment of deep brain stimulation for refractory depression.In the objective assessment of depression,the traditional depression scales are transformed into the interactive affective tasks.Based on the multimodal aBCI systems,multimodal physiological data including EEG,eye movement,etc,are collected simultaneously.Through deep learning,such as multimodal fusion and transfer learning,an objective assessment systems that can acurately distinguish depression states is established.In the treatment of refractory depression with deep brain stimulation,based on multimodal aBCI and reinforcement learning algorithms,the autonomously adjustment and personalization of the parameters are possible to improve the treatment outcomes.

关 键 词：抑郁症 多模态情感脑机接口 

分 类 号：R749.4[医药卫生—神经病学与精神病学]