CoRE: Concept-Reasoning Expansion for Continual Brain Lesion Segmentation

2026-04-28

AI Summary

This work addresses key challenges in continual learning for brain lesion segmentation in MRI, including catastrophic forgetting, limited model capacity, and the heterogeneity of pathological appearances across multimodal scans. To tackle these issues, the authors propose a Concept-Reasoning Expansion framework that aligns structured clinical concept knowledge with visual features. By integrating hierarchical concept representations, an expert routing mechanism, and dynamic model expansion, the method emulates clinical reasoning while avoiding excessive parameter growth. Evaluated on twelve sequential brain lesion MRI tasks, the approach achieves state-of-the-art performance, substantially enhancing few-shot transferability and decision interpretability. This study thus establishes an efficient and scalable knowledge foundation for continual learning in medical imaging.

Abstract

Accurate brain lesion segmentation in MRI is vital for effective clinical diagnosis and treatment planning. Due to high annotation costs and strict data privacy regulations, universal models require employing Continual Learning (CL) to adapt to evolving clinical tasks without losing previously acquired knowledge. However, existing CL paradigms often suffer from capacity limits or redundant parameter growth, and even advanced dynamic methods rely mostly on image-perception strategies that struggle to handle the substantial pathological and multimodal heterogeneity inherent in brain imaging. To address this issue, we propose Concept-Reasoning Expansion (CoRE) framework, which establishes a joint decision-making mechanism by integrating visual features with structured concepts. Through the alignment of image tokens with a hierarchical concept library, CoRE simulates clinical reasoning to guide both interpretable expert routing and demand-based model growth. This collaborative process ensures model evolution is grounded in clinical priors, preventing redundant parameter expansion while maximizing knowledge reuse. Extensive evaluations across 12 sequential brain lesion MRI tasks demonstrate that CoRE achieves state-of-the-art performance and provides a high knowledge starting point for efficient future adaptation. Its superior few-shot transferability and clinical interpretability further validate its effectiveness in managing non-stationary clinical data streams. Our code will be released soon.

Problem

• continual learning
• brain lesion segmentation
• pathological heterogeneity
• multimodal heterogeneity
• clinical interpretability

Innovation

• Continual Learning
• Concept-Reasoning Expansion
• Brain Lesion Segmentation
• Clinical Reasoning
• Knowledge Reuse