Research map / Continual learning / Class-incremental
Continual Segment: Towards a Single, Unified and Non-forgetting Continual Segmentation Model of 143 Whole-body Organs in CT Scans
Zhanghexuan Ji, Dazhou Guo, Puyang Wang, Ke Yan, Le Lu, Minfeng Xu, Qifeng Wang, Jia Ge, Mingchen Gao, Xianghua Ye, Dakai Jin
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| freeze the General Encoder when learning new datasets (ablation) | Continual segmentation framework · TotalSeg -> ChestOrgan (old task TotalSeg) | DSC (%): 51.42 -> 92.98 | Improved |
| freeze the General Encoder (effect on new task) (ablation) | Continual segmentation framework · TotalSeg -> HNOrgan (new task HNOrgan) | DSC (%): 84.35 -> 84.14 | No clear effect |
| use anomaly-aware merging instead of simple ensemble (ablation) | Output merging module · EsoOrgan | DSC (%): 80.22 -> 86.94 | Improved |
| use anomaly-aware merging instead of simple ensemble (ablation) | Output merging module · HNOrgan | DSC (%): 77.84 -> 83.97 | Improved |
| allow larger DSC drop tolerance in decoder pruning (1% to 5%) (ablation) | TotalSeg decoder · TotalSeg | relative decoder size: 0.43 -> 0.22 (DSC 92.98 -> 88.83) | Mixed |
| train General Encoder on ChestOrgan instead of TotalSegmentator (ablation) | Continual segmentation framework · Order A final results | DSC: <1% Dice drop | No clear effect |
| use frozen encoder with incrementally added decoders vs regularization-based CSS methods | Continual segmentation framework · 143 organs, Order A | mean DSC (%): 42.43 -> 88.74 | Improved |
Closest papers by meaning
- DoDNet: Learning To Segment Multi-Organ and Tumors From Multiple Partially Labeled Datasets (CVPR 2021 · similarity 0.90)
- Learning From Partially Labeled Data for Multi-Organ and Tumor Segmentation (TPAMI 2023 · similarity 0.90)
- MedCRP-CL: Continual Medical Image Segmentation via Bayesian Nonparametric Semantic Modality Discovery (ICML 2026 · similarity 0.89)
- CCQ: Cross-Class Query Network for Partially Labeled Organ Segmentation (AAAI 2023 · similarity 0.88)
- CAT: Coordinating Anatomical-Textual Prompts for Multi-Organ and Tumor Segmentation (NeurIPS 2024 · similarity 0.88)
- Towards a Comprehensive, Efficient and Promptable Anatomic Structure Segmentation Model Using 3D Whole-Body CT Scans (AAAI 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
