Modality-Agnostic Brain Lesion Segmentation with Privacy-aware Continual Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sadegheih, Yousef, Kumari, Pratibha, Merhof, Dorit
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912498817433600
author Sadegheih, Yousef
Kumari, Pratibha
Merhof, Dorit
author_facet Sadegheih, Yousef
Kumari, Pratibha
Merhof, Dorit
contents Traditional brain lesion segmentation models for multi-modal MRI are typically tailored to specific pathologies, relying on datasets with predefined modalities. Adapting to new MRI modalities or pathologies often requires training separate models, which contrasts with how medical professionals incrementally expand their expertise by learning from diverse datasets over time. Inspired by this human learning process, we propose a unified segmentation model capable of sequentially learning from multiple datasets with varying modalities and pathologies. Our approach leverages a privacy-aware continual learning framework that integrates a mixture-of-experts mechanism and dual knowledge distillation to mitigate catastrophic forgetting while not compromising performance on newly encountered datasets. Extensive experiments across five diverse brain MRI datasets and four dataset sequences demonstrate the effectiveness of our framework in maintaining a single adaptable model, capable of handling varying hospital protocols, imaging modalities, and disease types. Compared to widely used privacy-aware continual learning methods such as LwF, SI, EWC, MiB, and TED, our method achieves an average Dice score improvement of approximately 14%. Our framework represents a significant step toward more versatile and practical brain lesion segmentation models, with implementation available on \href{https://github.com/xmindflow/BrainCL}{GitHub}.
format Preprint
id arxiv_https___arxiv_org_abs_2503_20326
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Modality-Agnostic Brain Lesion Segmentation with Privacy-aware Continual Learning
Sadegheih, Yousef
Kumari, Pratibha
Merhof, Dorit
Image and Video Processing
Traditional brain lesion segmentation models for multi-modal MRI are typically tailored to specific pathologies, relying on datasets with predefined modalities. Adapting to new MRI modalities or pathologies often requires training separate models, which contrasts with how medical professionals incrementally expand their expertise by learning from diverse datasets over time. Inspired by this human learning process, we propose a unified segmentation model capable of sequentially learning from multiple datasets with varying modalities and pathologies. Our approach leverages a privacy-aware continual learning framework that integrates a mixture-of-experts mechanism and dual knowledge distillation to mitigate catastrophic forgetting while not compromising performance on newly encountered datasets. Extensive experiments across five diverse brain MRI datasets and four dataset sequences demonstrate the effectiveness of our framework in maintaining a single adaptable model, capable of handling varying hospital protocols, imaging modalities, and disease types. Compared to widely used privacy-aware continual learning methods such as LwF, SI, EWC, MiB, and TED, our method achieves an average Dice score improvement of approximately 14%. Our framework represents a significant step toward more versatile and practical brain lesion segmentation models, with implementation available on \href{https://github.com/xmindflow/BrainCL}{GitHub}.
title Modality-Agnostic Brain Lesion Segmentation with Privacy-aware Continual Learning
topic Image and Video Processing
url https://arxiv.org/abs/2503.20326