Mansour L Sina, Seguin Caio, Winkler Anderson M, Noble Stephanie, Zalesky Andrew
Department of Biomedical Engineering, The University of Melbourne, Melbourne, Victoria, Australia.
Melbourne Neuropsychiatry Centre, The University of Melbourne, Melbourne, Victoria, Australia.
Netw Neurosci. 2024 Oct 1;8(3):902-925. doi: 10.1162/netn_a_00375. eCollection 2024.
Functional magnetic resonance imaging (fMRI) studies most commonly use cluster-based inference to detect local changes in brain activity. Insufficient statistical power and disproportionate false-positive rates reportedly hinder optimal inference. We propose a structural connectivity-guided clustering framework, called topological cluster statistic (TCS), that enhances sensitivity by leveraging white matter anatomical connectivity information. TCS harnesses multimodal information from diffusion tractography and functional imaging to improve task fMRI activation inference. Compared to conventional approaches, TCS consistently improves power over a wide range of effects. This improvement results in a 10%-50% increase in local sensitivity with the greatest gains for medium-sized effects. TCS additionally enables inspection of underlying anatomical networks and thus uncovers knowledge regarding the anatomical underpinnings of brain activation. This novel approach is made available in the PALM software to facilitate usability. Given the increasing recognition that activation reflects widespread, coordinated processes, TCS provides a way to integrate the known structure underlying widespread activations into neuroimaging analyses moving forward.
功能磁共振成像（fMRI）研究最常使用基于簇的推断来检测大脑活动的局部变化。据报道，统计能力不足和假阳性率过高阻碍了最佳推断。我们提出了一种基于结构连通性的聚类框架，称为拓扑簇统计（TCS），它通过利用白质解剖连通性信息来提高灵敏度。TCS利用来自扩散张量成像和功能成像的多模态信息来改善任务fMRI激活推断。与传统方法相比，TCS在广泛的效应范围内持续提高了统计能力。这种改进导致局部灵敏度提高了10%-50%，对中等大小的效应增益最大。TCS还能够检查潜在的解剖网络，从而揭示有关大脑激活的解剖学基础的知识。这种新方法已在PALM软件中提供，以方便使用。鉴于人们越来越认识到激活反映了广泛的、协调的过程，TCS提供了一种方法，将广泛激活背后的已知结构整合到未来的神经成像分析中。
