Title
Diffusion-informed spatial smoothing of fMRI data in white matter using spectral graph filters
Abstract
•The BOLD signal in white matter has an anisotropic correlation structure that has not been accounted for in standard fMRI pipelines.•We present a graph-based spatial smoothing approach that yields filters adapted to the underlying diffusion MRI data of the subject.•The proposed smoothing method outperforms isotropic Gaussian filtering in simulated phantom experiments.•Single-subject and group activation maps from task fMRI data smoothed with the proposed approach manifest slender, streamline-like activation patterns that are generally absent when the data are smoothed with isotropic Gaussian filters.
Year
DOI
Venue
2021
10.1016/j.neuroimage.2021.118095
NeuroImage
Keywords
DocType
Volume
Functional MRI,Diffusion MRI,White matter,Graph signal processing,Anisotropy
Journal
237
ISSN
Citations 
PageRank 
1053-8119
1
0.35
References 
Authors
35
6
Name
Order
Citations
PageRank
David Abramian110.35
Martin Larsson212.38
Anders Eklund310.35
Iman Aganj419518.93
Carl-Fredrik Westin510.35
Hamid Behjat610.35