文章摘要

文章信息/Info

Title:: Application of regression algorithm in the auxiliary diagnosis of neuropsychiatric diseases

作者:: 黄家源1 ; 吴凯1; 2; 3; 4; 熊冬生2; 5; 邓泽亚6; 周静2; 3; 4; 5△; (1.华南理工大学生物医学科学与工程学院,广州 511442；2.华南理工大学国家人体组织功能重建工程技术研究中心，广州 510006；3.广东省精神疾病转化医学工程技术研究中心,广州 510370；4.广东省老年痴呆诊断与康复工程技术研究中心,广州 510500；5.华南理工大学材料科学与工程学院,广州 510006；6.广州双悠生物科技有限责任公司，广州 510535)

Author(s):: HUANG Jiayuan1; WU Kai1; 2; 3; 4; XIONG Dongsheng2; 5; DENG Zeya6; ZHOU Jing2; 3; 4; 5; （1.School of Biomedical Sciences and Engineering, South China University of Technology, Guangzhou 511442, China; 2.National Engineering Research Center for Tissue Restoration and Reconstruction,South China University of Technology,Guangzhou 510006; 3.Guangdong Engineering Technology Research Center for Translational Medicine of Mental Disorders, Guangzhou 510370; 4.Guangdong Engineering Technology Research Center for Diagnosis and Rehabilitation of Dementia, Guangzhou 510500;5.School of Material Science and Engineering, South China University of Technology, Guangzhou 510006;? 6.Brain Power System Co., Ltd, Guangzhou, 510535）

Keywords:: Neuropsychiatric disorders; Machine learning; Regression prediction; Clinical scoring; Brain age

摘要:: 大脑是人体神经中枢和高级神经活动的结构基础。回归算法可预测连续变量，有利于揭示神经精神疾病的神经病理机制，能够更精准地实现神经精神疾病的辅助诊断和预后预测。随着神经影像技术的发展，将回归算法应用于神经精神疾病临床评分及脑龄预测的研究日益增加。本文综述了回归算法在神经精神疾病辅助诊断中的研究进展，总结和分析了阿尔兹海默病、精神分裂症、帕金森病和脑龄等领域的最新研究结果，并指出当前研究存在样本量较少和重复性较差的不足，最后对多站点协同分析方案的可行性和深度学习方法的应用前景进行展望。

Abstract:: The brain is the human′s nerve center and the structural basis of high-level neural activities.Regression algorithm can predict continuous variables, which is more conducive to reveal the neuropathological mechanism of neuropsychiatric diseases, and can more accurately realize the auxiliary diagnosis and prognosis prediction of neuropsychiatric diseases.With the development of neuroimaging technology, the application of regression algorithms to neuropsychiatric disorders clinical scores and brain age prediction is increasing.We summarize the basic principles,processes and systematically, review the recent progresses of the predication of clinical scores of neuropsychiatric disorders and brain age, such as Alzheimer′s disease,schizophrenia and Parkinson′s disease. Moreover, several limitations are addressed, such as small sample size and poor reproducibility. Finally, the feasibility of multi-sites collaborative analysis scheme and the application prospect of deep learning methods are prospected.