Mapping Caudolenticular Gray Matter Bridges in the Human Brain Striatum Through Diffusion Magnetic Resonance Imaging and Tractography

ABSTRACT In primates, the putamen and the caudate nucleus are connected by ~1 mm‐thick caudolenticular gray matter bridges (CLGBs) interspersed between the white matter bundles of the internal capsule. Little is understood about the functional or microstructural properties of the CLGBs. In studies p...

Full description

Saved in:
Bibliographic Details
Published inHuman brain mapping Vol. 46; no. 8; pp. e70245 - n/a
Main Authors Little, Graham, Poirier, Charles, Bore, Arnaud, Parent, Martin, Petit, Laurent, Descoteaux, Maxime
Format Journal Article
LanguageEnglish
Published Hoboken, USA John Wiley & Sons, Inc 01.06.2025
Wiley
Subjects
Online AccessGet full text
ISSN1065-9471
1097-0193
1097-0193
DOI10.1002/hbm.70245

Cover

More Information
Summary:ABSTRACT In primates, the putamen and the caudate nucleus are connected by ~1 mm‐thick caudolenticular gray matter bridges (CLGBs) interspersed between the white matter bundles of the internal capsule. Little is understood about the functional or microstructural properties of the CLGBs. In studies proposing high resolution diffusion magnetic resonance imaging (dMRI) techniques, CLGBs have been qualitatively identified as an example of superior imaging quality; however, the microstructural properties of these structures have yet to be examined. In this study, it is demonstrated for the first time that dMRI is sensitive to an organized anisotropic signal oriented in the direction parallel to the CLGBs, suggesting that dMRI could be a useful imaging method for probing the microstructure of the CLGBs. To demonstrate the anisotropic diffusion signal is coherently organized along the extent of the CLGBs and to enable a subsequent CLGB microstructural measurement, a customized tractography seeding and filtering method is proposed that utilizes the shape of the human striatum (putamen + caudate nucleus) to reconstruct the CLGBs in 3D. The proposed seeding strategy seeds tractography streamlines outward and normal to the surface of a 3D model of the striatum such that reconstructed streamlines are more likely to follow the diffusion signal peaks aligned parallel to the CLGBs. The method is applied to three different diffusion datasets, namely a high resolution 760 μm isotropic diffusion dataset acquired on a single subject, the test–retest cohort included as part of the human connectome project (N = 44) with diffusion data acquired at 1.25 mm isotropic, and a locally acquired “clinical” test–retest dataset acquired at 2.0 mm isotropic (N = 24). Reconstructed CLGBs directly overlap expected gray matter regions in the human brain for all three datasets. In addition, the method is shown to accurately reconstruct CLGBs repeatedly across multiple test–retest cohorts. The tractography CLGB reconstructions are then used to extract a quantitative measurement of microstructure from a local model of the diffusion signal along the CLGBs themselves. This is the first work to comprehensively study the CLGBs in vivo using dMRI and presents techniques suitable for future human neuroscience studies targeting these structures. This work demonstrates that diffusion MRI is sensitive to an anisotropic signal oriented along the caudolenticular gray matter bridges (CLGBs) of the striatum. A tractography approach was developed to track this signal, reconstructing the CLGBs and demonstrating that the signal is present across the entirety of these gray matter structures.
Bibliography:Funding
This work was supported by Natural Sciences and Engineering Research Council of Canada; Unifying Neuroscience and Artificial Intelligence; Centre de recherche du Centre hospitalier Université de Sherbrooke; Funds de Recherche du Québec Nature et technologies; Université de Sherbrooke; Part of this research was supported by the NSERC Discovery grant and the Université de Sherbrooke Institutional Chair in Neuroinformatics from Pr Descoteaux; National Institute of Mental Health, grant RO1MH113257.
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
Funding: This work was supported by Natural Sciences and Engineering Research Council of Canada; Unifying Neuroscience and Artificial Intelligence; Centre de recherche du Centre hospitalier Université de Sherbrooke; Funds de Recherche du Québec Nature et technologies; Université de Sherbrooke; Part of this research was supported by the NSERC Discovery grant and the Université de Sherbrooke Institutional Chair in Neuroinformatics from Pr Descoteaux; National Institute of Mental Health, grant RO1MH113257.
ISSN:1065-9471
1097-0193
1097-0193
DOI:10.1002/hbm.70245