神经影像统计建模方法前沿与进展  

Frontiers and advances in neuroimaging statistical modeling methods

作　　者：任鹏 蒋宇超[1,2] 尤佳 程炜 REN Peng;JIANG Yuchao;YOU Jia;CHENG Wei(Institute of Science and Technology for Brain-Inspired Intelligence,Fudan University,Shanghai 200433,China;Key Laboratory of Computational Neuroscience and Brain Inspired Intelligence(Fudan University),Ministry of Education,Shanghai 200433,China;Department of Neurology,Huashan Hospital,Fudan University,Shanghai 200433,China)

机构地区：[1]复旦大学类脑智能科学与技术研究院,上海200433 [2]复旦大学计算神经科学与类脑智能教育部重点实验室,上海200433 [3]复旦大学附属华山医院神经内科,上海200433

出　　处：《肿瘤影像学》2024年第2期121-126,共6页Oncoradiology

摘　　要：神经影像统计建模是神经科学和医学领域的一个关键分支,本文对近年来神经影像统计建模领域的重要方法学进展进行述评。首先,本文介绍了认知解码模型,重点讨论了如何利用正交分解方法和表征相似性分析来解析神经影像数据中潜在的认知加工过程。其次,本文探讨了神经影像个体化建模方法,包括规范性建模和个体脑功能剖分,以及它们在精神疾病研究中的应用。随后,本文讨论了数据驱动的疾病进展模型,阐述了如何利用机器学习和统计学工具推断出疾病生物标志物随时间而变化的进展模式,以及该方法在神经退行性疾病等领域的应用。最后,本文分析了基于人工智能的神经影像建模方法及其在神经影像领域中的应用。Neuroimaging statistical modeling is a critical branch in the fields of neuroscience and medicine.This article provided an overview of recent methodological advances in neuroimaging statistical models.Firstly,the article introduced cognitive decoding models,focusing on how orthogonal decomposition methods and representational similarity analysis can be used to resolve the underlying cognitive processes in neuroimaging data.Secondly,the article discussed methods for individualized neuroimaging modeling,including normative modeling and individual brain functional parcellation,along with their applications in psychiatric research.Subsequently,the article explored data-driven disease progression models,elucidating how machine learning and statistical tools can be utilized to infer the progression patterns of disease biomarkers over time,and their applications in fields such as neurodegenerative diseases.Finally,the article reviewed artificial intelligence-based neuroimaging modeling methods,along with their applications in neuroimaging analysis.

关 键 词：神经影像 认知建模 个体化建模 疾病进展模型 人工智能 

分 类 号：R739.4[医药卫生—肿瘤] R445.2[医药卫生—临床医学]