Research map / Detection and segmentation / Open-set and few-shot detection
Concept-to-Pixel: Prompt-Free Universal Medical Image Segmentation
Haoyun Chen, Fenghe Tang, Wenxin Ma, S Kevin 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 |
|---|---|---|---|
| add Semantic Tokens (ablation) | baseline without tokens · 8-dataset joint benchmark (average) | Dice (%): 82.38 -> 87.15 | Improved |
| add Geometry Tokens (ablation) | baseline without tokens · 8-dataset joint benchmark (average) | Dice (%): 82.38 -> 87.31 | Improved |
| replace static head with dynamic head (no tokens) (ablation) | baseline · 8-dataset joint benchmark (average) | Dice (%): 82.38 -> 87.19 | Improved |
| add Semantic Tokens to Geometry Tokens + dynamic head (ablation) | C2P without SEM · 8-dataset joint benchmark (average) | Dice (%): 87.29 -> 87.79 | Improved |
| add Geometry Tokens to Semantic Tokens + dynamic head (ablation) | C2P without GEO · 8-dataset joint benchmark (average) | Dice (%): 87.43 -> 87.79 | Improved |
| add dynamic head to Geometry Tokens (static head) (ablation) | GEO-only model · 8-dataset joint benchmark (average) | Dice (%): 87.31 -> 87.29 | No clear effect |
| add standard TTA at inference (ablation) | C2P without inference strategy · 8-dataset joint benchmark (average) | Dice (%): 87.79 -> 88.16 | Improved |
| replace standard TTA with Geometry-Aware Consistency (ablation) | C2P with standard TTA · 8-dataset joint benchmark (average) | Dice (%): 88.16 -> 88.22 | Improved |
| evaluate C2P instead of Spider (64-ref) zero-shot | Spider · ACDC (zero-shot) | Dice (%): 63.26 -> 71.87 | Improved |
| evaluate C2P instead of Spider (64-ref) zero-shot | Spider · ISBI EM (zero-shot, unseen modality) | Dice (%): 64.63 -> 26.79 | Hurt |
| evaluate C2P instead of Spider (64-ref) zero-shot | Spider · Montgomery X-Ray (zero-shot) | Dice (%): 83.89 -> 87.60 | Improved |
| evaluate C2P instead of Spider (64-ref) zero-shot | Spider · BUSBRA (zero-shot) | Dice (%): 85.16 -> 83.39 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- Spider — Spider: A Unified Framework for Context-dependent Concept Segmentation (ICML 2024) · better on 1 of 1 shared result
- UniverSeg — UniverSeg: Universal Medical Image Segmentation (ICCV 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Training Like a Medical Resident: Context-Prior Learning Toward Universal Medical Image Segmentation (CVPR 2024 · similarity 0.88)
- UniPro: Unified Multi-Mode Medical Image Segmentation from 2D Images to 3D Volumes via Propagation (NeurIPS 2026 · similarity 0.87)
- CAT: Coordinating Anatomical-Textual Prompts for Multi-Organ and Tumor Segmentation (NeurIPS 2024 · similarity 0.87)
- S2S2: Semantic Stacking for Robust Semantic Segmentation in Medical Imaging (AAAI 2025 · similarity 0.87)
- RADAR: Text-Guided Medical Image Segmentation via Residual Aggregation and Dense Alignment Representations (NeurIPS 2026 · similarity 0.87)
- Show and Segment: Universal Medical Image Segmentation via In-Context Learning (CVPR 2025 · similarity 0.86)
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.
