A Continual Learning-driven Model for Accurate and Generalizable Segmentation of Clinically Comprehensive and Fine-grained Whole-body Anatomies in CT
🤖 AI Summary
Addressing the scarcity of large-scale fully annotated CT datasets and the poor generalizability of segmentation models across imaging devices, acquisition phases, and disease conditions, this paper proposes CL-Net—the first continual learning framework for whole-body anatomical segmentation supporting 235 structures. Methodologically, it introduces a capacity-adaptive continual learning architecture integrating a universal encoder with pruned, task-specific decoders to mitigate catastrophic forgetting. It further incorporates Elastic Weight Consolidation (EWC) regularization, multi-task decoupled training, knowledge distillation, and a federated multi-source CT alignment strategy. Evaluated on 13,952 multi-center CT scans, CL-Net achieves superior performance over an ensemble of 36 nnUNet models—using only 5% of their parameters—and significantly outperforms SAM-style medical foundation models. The framework delivers high accuracy, strong cross-domain generalization, and seamless incremental learning capability for comprehensive whole-body anatomical parsing.
Application Category
📝 Abstract
Precision medicine in the quantitative management of chronic diseases and oncology would be greatly improved if the Computed Tomography (CT) scan of any patient could be segmented, parsed and analyzed in a precise and detailed way. However, there is no such fully annotated CT dataset with all anatomies delineated for training because of the exceptionally high manual cost, the need for specialized clinical expertise, and the time required to finish the task. To this end, we proposed a novel continual learning-driven CT model that can segment complete anatomies presented using dozens of previously partially labeled datasets, dynamically expanding its capacity to segment new ones without compromising previously learned organ knowledge. Existing multi-dataset approaches are not able to dynamically segment new anatomies without catastrophic forgetting and would encounter optimization difficulty or infeasibility when segmenting hundreds of anatomies across the whole range of body regions. Our single unified CT segmentation model, CL-Net, can highly accurately segment a clinically comprehensive set of 235 fine-grained whole-body anatomies. Composed of a universal encoder, multiple optimized and pruned decoders, CL-Net is developed using 13,952 CT scans from 20 public and 16 private high-quality partially labeled CT datasets of various vendors, different contrast phases, and pathologies. Extensive evaluation demonstrates that CL-Net consistently outperforms the upper limit of an ensemble of 36 specialist nnUNets trained per dataset with the complexity of 5% model size and significantly surpasses the segmentation accuracy of recent leading Segment Anything-style medical image foundation models by large margins. Our continual learning-driven CL-Net model would lay a solid foundation to facilitate many downstream tasks of oncology and chronic diseases using the most widely adopted CT imaging.
Problem
Research questions and friction points this paper is trying to address.
Accurate segmentation of whole-body anatomies in CT scans.
Overcoming lack of fully annotated CT datasets for training.
Dynamic learning to segment new anatomies without forgetting.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Continual learning-driven CT segmentation model
Dynamic expansion for new anatomy segmentation
Universal encoder with multiple optimized decoders
🔎 Similar Papers
No similar papers found.
