guest ::
| Home > Publications Database > Vessel distance mapping reveals mesoscopic arterial and venous patterns in the putamen, v1 |
| Home > Publications Database > Vessel distance mapping reveals mesoscopic arterial and venous patterns in the putamen, v1 |
| Preprint | DZNE-2025-00205 |
; ; ; ; ; ;
2023
Zenodo
This record in other databases:
Please use a persistent id in citations: doi:10.5281/zenodo.8098900 doi:10.5281/ZENODO.8098900
Abstract: Purpose: While vessel atlases, i.e. building one global mean representative vascular model, and vessel patterns, i.e. finding distinct differences in how a structure is vascularized across subjects, have gained momentum recently, both concepts are treated as two separated entities currently. The aim of this study was to bridge vessel atlases and patterns by identifying distinct arterial and venous vascular patterns in the human putamen at the mesoscopic scale using MRI. Methods: High resolution MRI of 20 healthy subjects (40 hemispheres) is leveraged to compute arterial and venous segmentations, respectively. The recently introduced Vessel Distance Mapping (VDM) framework was combined with so-called vascular fingerprints to find arterial and venous patterns in an unsupervised manner, respectively. As a complementary approach, qualitative pattern descriptions and expert ratings were generated. The patterns found are compared with each other and to the global mean in native image space and after spatial normalization to MNI space, respectively. Results: Automatic clustering and expert rating found similar vessel patterns in the putamen, with arterial and venous patterns following a unimodal and bimodal distribution, respectively. When neglecting patterns with single putamen assigned, the automatic clustering found 4 arterial and 3 venous, while the expert rating found 3 arterial and 4 venous patterns. While the patterns showed distinct differences w.r.t. each other, arterial vascularization in the putamen is well described by one global mean with unimodal distributed subject-specific variations. Venous pattern showed a bimodal distribution with either a lower or higher vascularization then the global mean, limiting the representative character of a single global representation as commonly used in vessel atlases. Conclusion: To you best knowledge, this is the first description of distinctively different vessel patterns in the human putamen using MRI. Further, we provide tools to investigate the vasculature beyond global averages, i.e. by identifying vessel patterns using either unsupervised clustering or expert ratings. Future applications included assessment of vessel biases in layer-specific function MRI and vascular resistance and resilience mechanisms in pathologies. Overall, the study advocates the use of vessel pattern-specifics atlases which become more relevant with increasing imaging resolutions as vessel patterns become more heterogeneous and explicit vessel co-registration is no longer possible.
Keyword(s): Vessel distance mapping ; Magnetic Resonance Imaging ; ultra-high field ; vessel imaging ; vasculature
; 
|
The record appears in these collections: |
ZIP