A Unified Framework for Joint Detection of Lacunes and Enlarged Perivascular Spaces

TL;DR

This paper introduces a novel, unified deep learning framework for simultaneous detection of lacunes and enlarged perivascular spaces in brain MRI, addressing class imbalance and feature interference to improve accuracy.

Contribution

The proposed morphology-decoupled framework with cross-task attention, mixed-supervision, and anatomical calibration advances joint detection of CSVD markers beyond existing methods.

Findings

Achieved state-of-the-art lacune detection precision of 71.1%.

Improved F1-score to 62.6%, surpassing previous methods.

Demonstrated robustness on large external cohort.

Abstract

Cerebral small vessel disease (CSVD) markers, specifically enlarged perivascular spaces (EPVS) and lacunae, present a unique challenge in medical image analysis due to their radiological mimicry. Standard segmentation networks struggle with feature interference and extreme class imbalance when handling these divergent targets simultaneously. To address these issues, we propose a morphology-decoupled framework where Zero-Initialized Gated Cross-Task Attention exploits dense EPVS context to guide sparse lacune detection. Furthermore, biological and topological consistency are enforced via a mixed-supervision strategy integrating Mutual Exclusion and Centerline Dice losses. Finally, we introduce an Anatomically-Informed Inference Calibration mechanism to dynamically suppress false positives based on tissue semantics. Extensive 5-folds cross-validation on the VALDO 2021 dataset (N=40) demonstrates state-of-the-art performance, notably surpassing task winners in lacunae detection precision (71.1%, p=0.01) and F1-score (62.6%, p=0.03). Furthermore, evaluation on the external EPAD cohort (N=1762) confirms the model's robustness for large-scale population studies. Code will be released upon acceptance.