3D Classification of Paramagnetic Rim Lesions in Multiple Sclerosis via Asymmetric QSM-FLAIR Modeling

2026-06-15

AI Summary

This study addresses the challenges of automatic detection of paramagnetic rim lesions (Rim⁺) in multiple sclerosis, which are hindered by data scarcity, multimodal heterogeneity, and extreme class imbalance. The authors propose a 3D multimodal deep learning framework that innovatively employs quantitative susceptibility mapping (QSM) as the primary modality and FLAIR for structural context through an asymmetric fusion mechanism. To enhance model robustness under limited data, they integrate self-supervised multimodal pretraining with contrastive regularized fine-tuning. Evaluated on clinical data from 88 MS patients, the method outperforms existing approaches in lesion-level Rim⁺/Rim⁻ classification, demonstrating its effectiveness in enabling automated identification of chronic active lesions.

Abstract

Paramagnetic rim lesions (Rim$^+$) identified on susceptibility-sensitive MRI have recently emerged as a specific biomarker of chronic active inflammation in Multiple Sclerosis (MS) and are associated with long-term disability progression. However, susceptibility imaging and expert interpretation remain limited to specialized centers, visual assessment is time-consuming and variable, and the low prevalence of Rim$^+$ lesions poses severe class imbalance challenges for automated analysis. We propose a 3D multimodal deep learning framework for lesion-level Rim$^+$/Rim$^-$ classification from Quantitative Susceptibility Mapping (QSM) and FLAIR MRI. The architecture explicitly models modality asymmetry by treating QSM as the primary susceptibility-driven signal and conditioning it with FLAIR-derived structural context. To improve robustness under limited data, we employ self-supervised multimodal pretraining followed by supervised fine-tuning with contrastive regularization. The method was evaluated on a clinically acquired cohort of 88 people with MS with expert lesion annotations as reference standard. Results highlight improved performance compared to prior architectures, supporting the effectiveness of asymmetric multimodal modeling for automated chronic active lesion identification.

Problem

• Paramagnetic rim lesions
• Multiple Sclerosis
• Class imbalance
• Susceptibility imaging
• Automated lesion classification

Innovation

• asymmetric multimodal modeling
• paramagnetic rim lesions
• QSM-FLAIR fusion
• self-supervised pretraining
• contrastive regularization