CenSynCMB: Centre Maps and Physics-Guided Synthesis for Microbleed Detection

2026-07-06

AI Summary

Cerebral microbleeds (CMBs) are extremely challenging to detect automatically in MRI due to their small size, sparse distribution, and frequent confusion with vessels, calcifications, and imaging artifacts. To address this, this work proposes the CenSynCMB framework, which innovatively integrates a 3D Attention U-Net with center-map-guided supervision, a false-negative reweighting strategy, and a physics-informed synthetic data generation method to produce realistic CMBs and hard negative samples. This approach mitigates annotation scarcity without data leakage and substantially enhances the model’s ability to discriminate subtle lesions from confounding structures. The method achieves state-of-the-art lesion-level performance with an F1 score of 74.3% on VALDO Task 2 and demonstrates superior generalization on the external AIBL SWI dataset, attaining the highest recall of 88.5% and an F1 score of 65.0%, outperforming existing approaches.

Abstract

Cerebral microbleeds (CMBs) are MRI markers of small vessel disease and the microbleed component of amyloid related imaging abnormalities (ARIA-H), but their small size, sparsity, and similarity to vessels, calcification-like foci, and artefacts make automated detection difficult. We propose CenSynCMB, a centre-guided and mimic-aware framework combining a 3D Attention U-Net, auxiliary centre-map supervision, false-negative-driven reweighting, and fold-wise physics-guided synthesis of positive CMBs and labelled hard negatives. Synthetic data expose the detector to compact lesions and common mimics without validation or test leakage. On VALDO Task 2, CenSynCMB achieved the best local-comparison lesion-level F1 (74.3%, p = 0.020); on external AIBL SWI, it achieved the highest local-comparison recall (88.5%, p = 0.0058) and F1 (65.0%, p = 0.0016). Together, these results support scalable CMB candidate extraction in large, unlabelled MRI cohorts, while highlighting cohort-specific calibration as the next step toward reliable burden estimation.

Problem

• cerebral microbleeds
• automated detection
• MRI
• small vessel disease
• ARIA-H

Innovation

• centre-map supervision
• physics-guided synthesis
• false-negative-driven reweighting
• 3D Attention U-Net
• hard negative mining