MedNet-PVS: A MedNeXt-Based Deep Learning Model for Automated Segmentation of Perivascular Spaces

2025-08-27

AI Summary

Manual segmentation of perivascular spaces (PVS) is time-consuming and suffers from poor inter-rater reliability, while existing deep learning models exhibit limited generalizability across MRI sequences and multi-center datasets. Method: We propose MedNeXt-L-k5—a 3D encoder-decoder network inspired by the Transformer architecture—to enable robust, automatic PVS segmentation across MRI modalities (T1w/T2w) and multi-center data. Evaluation employs 5-fold and leave-one-center-out cross-validation. Results: On HCP-Aging T2w data, our method achieves a state-of-the-art voxel-level Dice score of 0.88±0.06, matching human annotator consistency; it further demonstrates superior cluster-level stability across centers compared to prior approaches. This work delivers the first automated tool for PVS quantification that simultaneously achieves high accuracy, strong generalizability, and cross-sequence compatibility—enabling scalable, multi-center neuroimaging studies.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Enlarged perivascular spaces (PVS) are increasingly recognized as biomarkers of cerebral small vessel disease, Alzheimer's disease, stroke, and aging-related neurodegeneration. However, manual segmentation of PVS is time-consuming and subject to moderate inter-rater reliability, while existing automated deep learning models have moderate performance and typically fail to generalize across diverse clinical and research MRI datasets. We adapted MedNeXt-L-k5, a Transformer-inspired 3D encoder-decoder convolutional network, for automated PVS segmentation. Two models were trained: one using a homogeneous dataset of 200 T2-weighted (T2w) MRI scans from the Human Connectome Project-Aging (HCP-Aging) dataset and another using 40 heterogeneous T1-weighted (T1w) MRI volumes from seven studies across six scanners. Model performance was evaluated using internal 5-fold cross validation (5FCV) and leave-one-site-out cross validation (LOSOCV). MedNeXt-L-k5 models trained on the T2w images of the HCP-Aging dataset achieved voxel-level Dice scores of 0.88+/-0.06 (white matter, WM), comparable to the reported inter-rater reliability of that dataset, and the highest yet reported in the literature. The same models trained on the T1w images of the HCP-Aging dataset achieved a substantially lower Dice score of 0.58+/-0.09 (WM). Under LOSOCV, the model had voxel-level Dice scores of 0.38+/-0.16 (WM) and 0.35+/-0.12 (BG), and cluster-level Dice scores of 0.61+/-0.19 (WM) and 0.62+/-0.21 (BG). MedNeXt-L-k5 provides an efficient solution for automated PVS segmentation across diverse T1w and T2w MRI datasets. MedNeXt-L-k5 did not outperform the nnU-Net, indicating that the attention-based mechanisms present in transformer-inspired models to provide global context are not required for high accuracy in PVS segmentation.

Problem

• Automating segmentation of perivascular spaces in MRI scans
• Addressing limited generalization across diverse clinical datasets
• Overcoming time-consuming manual segmentation with moderate reliability

Innovation

• Adapted MedNeXt-L-k5 transformer-inspired 3D network
• Trained models on both homogeneous and heterogeneous datasets
• Used cross-validation methods for performance evaluation