Activated fibers: Fiber-centered activation detection in task-based FMRI

Jinglei Lv, Lei Guo, Kaiming Li, Xintao Hu, Dajiang Zhu, Junwei Han, Tianming Liu

科研成果: 书/报告/会议事项章节会议稿件同行评审

10 引用 (Scopus)

摘要

In task-based fMRI, the generalized linear model (GLM) is widely used to detect activated brain regions. A fundamental assumption in the GLM model for fMRI activation detection is that the brain's response, represented by the blood-oxygenation level dependent (BOLD) signals of volumetric voxels, follows the shape of stimulus paradigm. Based on this same assumption, we use the dynamic functional connectivity (DFC) curves between two ends of a white matter fiber, instead of the BOLD signal, to represent the brain's response, and apply the GLM to detect Activated Fibers (AFs). Our rational is that brain regions connected by white matter fibers tend to be more synchronized during stimulus intervals than during baseline intervals. Therefore, the DFC curves for fibers connecting active brain regions should be positively correlated with the stimulus paradigm, which is verified by our extensive experiments using multimodal task-based fMRI and diffusion tensor imaging (DTI) data. Our results demonstrate that the detected AFs connect not only most of the activated brain regions detected via traditional voxel-based GLM method, but also many other brain regions, suggesting that the voxel-based GLM method may be too conservative in detecting activated brain regions.

源语言英语
主期刊名Information Processing in Medical Imaging - 22nd International Conference, IPMI 2011, Proceedings
574-587
页数14
DOI
出版状态已出版 - 2011
活动22nd International Conference on Information Processing in Medical Imaging, IPMI 2011 - Kloster Irsee, 德国
期限: 3 7月 20118 7月 2011

出版系列

姓名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
6801 LNCS
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议22nd International Conference on Information Processing in Medical Imaging, IPMI 2011
国家/地区德国
Kloster Irsee
时期3/07/118/07/11

指纹

探究 'Activated fibers: Fiber-centered activation detection in task-based FMRI' 的科研主题。它们共同构成独一无二的指纹。

引用此