心理科学进展2025，Vol.33Issue(6)：887-904,18.DOI:10.3724/SP.J.1042.2025.0887
融合机器学习技术的阈下抑郁神经生理机制及干预
Neurophysiological mechanisms and interventions of subthreshold depression by integrating machine learning techniques
摘要
Abstract
Major Depressive Disorder(MDD)poses a substantial threat to national mental health.Subthreshold depression,serving as a crucial prodromal stage of MDD,is of great value for investigating the neurophysiological features and its dynamic developmental patterns,as well as their potential for improving prediction of MDD onset.Past research is limited in treating MDD as a static,singular diagnostic entity.The current research,grounded in complex dynamic systems theory,explores multi-temporal and multi-modal machine learning techniques to explore the intricate relationships between subthreshold depressive symptoms and neurophysiological characteristics,as well as to identify key predictive factors.Additionally,through longitudinal tracking and neurodynamic network modeling,the study investigates attractor states and their predictive capacity for subsequent MDD onset and characteristic transitions.Additionally,the current study explores the preventive efficacy of cognitive behavioral therapy for subthreshold depression and the predictive role of attractor states.The research aims to clarify the neurophysiological features and its dynamic developmental patterns of subthreshold depression,hoping to inform the development of effective earlv screening and selective prevention strategies of MDD.关键词
阈下抑郁/吸引子状态/认知行为疗法/预防性干预/多模态机器学习Key words
subthreshold depression/attractors/cognitive-behavioral therapy/preventive intervention/multi-modal machine learning分类
医药卫生引用本文复制引用
刘永进,杨雪,杜欣欣,嵇文麒,臧寅垠,官锐园,宋森,钱铭怡,牟文婷..融合机器学习技术的阈下抑郁神经生理机制及干预[J].心理科学进展,2025,33(6):887-904,18.基金项目
国家自然科学基金(批准号:U2336214)资助. （批准号:U2336214）
