Research map / Detection and segmentation / Open-set and few-shot detection
AbdomenAtlas-8K: Annotating 8,000 CT Volumes for Multi-Organ Segmentation in Three Weeks
Chongyu Qu, Tiezheng Zhang, Hualin Qiao, Jie Liu, Yucheng Tang, Alan Yuille, Zongwei Zhou
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 |
|---|---|---|---|
| train segmentation model on AbdomenAtlas-8K instead of JHH private data (SwinUNETR) | SwinUNETR · JHH test | mDSC (%): 84.8 -> 86.5 | Improved |
| train on AbdomenAtlas-8K instead of JHH (UNETR) | UNETR · JHH test | mDSC (%): 78.6 -> 86.6 | Improved |
| train on AbdomenAtlas-8K instead of JHH (U-Net) | U-Net · JHH test | mDSC (%): 84.8 -> 87.3 | Improved |
| train on AbdomenAtlas-8K instead of JHH (SegResNet) | SegResNet · JHH test | mDSC (%): 87.6 -> 87.0 | No clear effect |
| train on AbdomenAtlas-8K instead of JHH (SwinUNETR) | SwinUNETR · JHH test | mNSD (%): 66.5 -> 60.8 | Hurt |
| train on AbdomenAtlas-8K instead of JHH (UNETR) | UNETR · JHH test | mNSD (%): 53.9 -> 59.5 | Improved |
| train on AbdomenAtlas-8K instead of JHH (U-Net) | U-Net · JHH test | mNSD (%): 64.2 -> 61.0 | Hurt |
| train on AbdomenAtlas-8K instead of JHH (SegResNet) | SegResNet · JHH test | mNSD (%): 64.9 -> 60.4 | Hurt |
| fine-tune with annotator-revised labels over active learning rounds (aorta) (ablation) | Active learning pipeline · JHH 1000 CTs | DSC (%): 72.3 -> 83.7 | Improved |
| fine-tune with annotator-revised labels over active learning rounds (postcava) (ablation) | Active learning pipeline · JHH 1000 CTs | DSC (%): 76.1 -> 78.6 | Improved |
| prioritize top 5% of CTs by attention map for human revision instead of annotating every organ manually | Active learning annotation · 8,000 CT volumes | annotation time: 30.8 years -> three weeks | Improved |
| manual per-CT annotation to attention-guided revision | Active learning annotation · per CT volume | annotation speedup: 60×8/15=32 | Improved |
Closest papers by meaning
- AbdomenCT-1K: Is Abdominal Organ Segmentation a Solved Problem? (TPAMI 2021 · similarity 0.92)
- CAT: Coordinating Anatomical-Textual Prompts for Multi-Organ and Tumor Segmentation (NeurIPS 2024 · similarity 0.87)
- Touchstone Benchmark: Are We on the Right Way for Evaluating AI Algorithms for Medical Segmentation? (NeurIPS 2024 · similarity 0.87)
- Versatile Medical Image Segmentation Learned from Multi-Source Datasets via Model Self-Disambiguation (CVPR 2024 · similarity 0.86)
- MultiOrg: A Multi-rater Organoid-detection Dataset (NeurIPS 2024 · similarity 0.85)
- PanTS: The Pancreatic Tumor Segmentation Dataset (NeurIPS 2025 · similarity 0.85)
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.
