Young versus older subject diffusion magnetic resonance imaging data for virtual white matter lesion tractography

White matter hyperintensity (WMH) lesions on T2 fluid‐attenuated inversion recovery (FLAIR) magnetic resonance imaging (MRI) and changes in adjacent normal‐appearing white matter can disrupt computerized tract reconstruction and result in inaccurate measures of structural brain connectivity. The vir...

Full description

Saved in:
Bibliographic Details
Published in:Human brain mapping 2023-07, Vol.44 (10), p.3943-3953
Main Authors: Taghvaei, Mohammad, Cook, Philip, Sadaghiani, Shokufeh, Shakibajahromi, Banafsheh, Tackett, William, Dolui, Sudipto, De, Debarun, Brown, Christopher, Khandelwal, Pulkit, Yushkevich, Paul, Das, Sandhitsu, Wolk, David A., Detre, John A.
Format: Article
Language:English
Subjects:
Aged
Aging - pathology
Brain - diagnostic imaging
Brain - pathology
brain connectivity
Corpus callosum
Data acquisition
Diffusion Magnetic Resonance Imaging - methods
Fornix
Humans
Image reconstruction
Lesions
Leukoaraiosis - pathology
Life span
Magnetic resonance imaging
Magnetic Resonance Imaging - methods
Masks
Medical imaging
Neural networks
Neuroimaging
Substantia alba
tractography
virtual lesion
White Matter - diagnostic imaging
White Matter - pathology
white matter hyperintensity
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:White matter hyperintensity (WMH) lesions on T2 fluid‐attenuated inversion recovery (FLAIR) magnetic resonance imaging (MRI) and changes in adjacent normal‐appearing white matter can disrupt computerized tract reconstruction and result in inaccurate measures of structural brain connectivity. The virtual lesion approach provides an alternative strategy for estimating structural connectivity changes due to WMH. To assess the impact of using young versus older subject diffusion MRI data for virtual lesion tractography, we leveraged recently available diffusion MRI data from the Human Connectome Project (HCP) Lifespan database. Neuroimaging data from 50 healthy young (39.2 ± 1.6 years) and 46 healthy older (74.2 ± 2.5 years) subjects were obtained from the publicly available HCP‐Aging database. Three WMH masks with low, moderate, and high lesion burdens were extracted from the WMH lesion frequency map of locally acquired FLAIR MRI data. Deterministic tractography was conducted to extract streamlines in 21 WM bundles with and without the WMH masks as regions of avoidance in both young and older cohorts. For intact tractography without virtual lesion masks, 7 out of 21 WM pathways showed a significantly lower number of streamlines in older subjects compared to young subjects. A decrease in streamline count with higher native lesion burden was found in corpus callosum, corticostriatal tract, and fornix pathways. Comparable percentages of affected streamlines were obtained in young and older groups with virtual lesion tractography using the three WMH lesion masks of increasing severity. We conclude that using normative diffusion MRI data from young subjects for virtual lesion tractography of WMH is, in most cases, preferable to using age‐matched normative data. The major drawback of using target diffusion magnetic resonance imaging (MRI) data from older subjects for virtual lesion tractography is the detection of fewer streamlines. For some white matter (WM) tracts, the presence of WM hyperintensity (WMH) in older subjects further contributes to an underestimation of actual tract disconnection. We conclude that using diffusion MRI data of young subjects for virtual lesion tractography of WMH is, in most cases, preferable to using age‐matched normative diffusion MRI data.
ISSN:1065-9471
1097-0193
1097-0193
DOI:10.1002/hbm.26326