aiqbench analysis
MedCRP-CL addresses the critical challenge of catastrophic forgetting in medical image segmentation by dynamically discovering task structures using Bayesian nonparametric methods without requiring raw data replay. By leveraging clinical text prompts to infer semantic modalities, this framework enables efficient knowledge transfer between similar tasks while isolating parameters for dissimilar ones, significantly reducing parameter count and memory overhead. This approach represents a meaningful advancement in continual learning architectures specifically tailored for heterogeneous medical imaging datasets.
MedCRP-CL: Continual Medical Image Segmentation via Bayesian Nonparametric Semantic Modality Discovery
arXiv:2605.20297v2 Announce Type: replace-cross Abstract: Medical image segmentation faces a fundamental challenge in continual learning: data arrives sequentially from heterogeneous sources, yet effective continual learning requires discovering which tasks share sufficient structure to benefit from joint learning. Existing methods either apply uniform constraints across all tasks, causing catastrophic forgetting when tasks conflict, or require predefined task groupings that cannot anticipate future task diversity. We introduce MedCRP-CL, a framework that performs online task structure discovery and structure-aware continual learning. Leveraging the Chinese Restaurant Process (CRP), our method dynamically infers task groupings from clinical text prompts as tasks arrive, without requiring predefined cluster counts or access to future tasks. We term these discovered groupings semantic modalities, as they capture finer-grained structure than physical imaging modalities by integrating anatomical region and pathological context. Guided by this discovered structure, we maintain semantic modality-specific LoRA adapters regularized by intra-modality EWC, ensuring parameter isolation across dissimilar task groups while facilitating knowledge transfer within similar ones. The framework is also replay-free, storing only aggregate statistics rather than raw patient data. Experiments on 16 medical segmentation tasks across four imaging modalities demonstrate that MedCRP-CL achieves 73.3% Dice score with only 4.1% forgetting, outperforming the best baseline by 8.0% while requiring 6$\times$ fewer parameters. Code is available at https://github.com/zygao930/MedCRP-CL.
Abstract:Medical image segmentation faces a fundamental challenge in continual learning: data arrives sequentially from heterogeneous sources, yet effective continual learning requires discovering which tasks share sufficient structure to benefit from joint learning. Existing methods either apply uniform constraints across all tasks, causing catastrophic forgetting when tasks conflict, or require predefined task groupings that cannot anticipate future task diversity. We introduce MedCRP-CL, a framework that performs online task structure discovery and structure-aware continual learning. Leveraging the Chinese Restaurant Process (CRP), our method dynamically infers task groupings from clinical text prompts as tasks arrive, without requiring predefined cluster counts or access to future tasks. We term these discovered groupings semantic modalities, as they capture finer-grained structure than physical imaging modalities by integrating anatomical region and pathological context. Guided by this discovered structure, we maintain semantic modality-specific LoRA adapters regularized by intra-modality EWC, ensuring parameter isolation across dissimilar task groups while facilitating knowledge transfer within similar ones. The framework is also replay-free, storing only aggregate statistics rather than raw patient data. Experiments on 16 medical segmentation tasks across four imaging modalities demonstrate that MedCRP-CL achieves 73.3% Dice score with only 4.1% forgetting, outperforming the best baseline by 8.0% while requiring 6$\times$ fewer parameters. Code is available at this https URL.
Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
arXiv:2605.20297 [cs.CV]
(or arXiv:2605.20297v2 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2605.20297 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 305 words
Related coverage
++nnU-Net: Scaling nnU-Net with Prefix-Based Data Augmentation
arXiv · Jun 10, 2026
[AINews] not much happened today
Latent Space · Sep 10, 2026
Biomedical Imaging's Real Bottleneck Is the Data, Not the Model
Databricks · Oct 8, 2026
MedLVR: Latent Visual Reasoning for Reliable Medical Visual Question Answering
arXiv · Jul 21, 2026