tsta

Home / سی و دومین کنفرانس ملی و دهمین کنفرانس بین المللی مهندسی زیست پزشکی ایران

Quantitative Mapping of Perivascular Spaces Across MRI Modalities Using Vesselness Filtering and Morphometric Analysis

Authors :

Razieh Salesi¹ Hamid Soltanian-Zadeh²

1- School of Electrical and Computer Engineering, University of Tehran 2- School of Electrical and Computer Engineering, University of Tehran

Keywords :

Glymphatic،Perivascular Space (PVS)،Vesselness Filtering،Enhanced Perivascular Contrast (EPC)،Human Connectome Project (HCP)،Structural MRI

Abstract :

Perivascular spaces (PVS), fluid-filled structures around cerebral vessels, are key markers of glymphatic function. We compared T1-weighted (T1w), T2-weighted (T2w), and Enhanced PVS Contrast (EPC) MRI for automated PVS segmentation and quantification. We used high-resolution MRI data from 50 healthy Human Connectome Project (HCP) participants. We applied denoising and vesselness filtering to enhance PVS structures. PVS masks were generated and refined using thresholding and skeletonization. We extracted volume, count, diameter, and contrast metrics for comparison. T2w yielded the highest PVS volume and count; EPC showed the largest diameters. All modalities differed significantly across most metrics (p < 0.001). Skeletonization reduced volume and count but preserved relative differences. EPC consistently produced the largest diameter estimates, both before and after skeletonization, suggesting improved structural delineation of central PVS regions. While contrast-based metrics showed weaker and less consistent differences, mean diameter and volume remained the most discriminative. T2w and EPC offer complementary strengths, with EPC excelling in PVS detail.