Research map / Label-efficient and robust learning / Test-time adaptation
Progressive Test Time Energy Adaptation for Medical Image Segmentation
Xiaoran Zhang, Byung-Woo Hong, Hyoungseob Park, Daniel H. Pak, Anne-Marie Rickmann, Lawrence H. Staib, James S. Duncan, Alex Wong
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 progressive energy-based TTA | pretrained UNet · ACDC to LVQuant LV | DSC (%): 58.98 -> 76.93 | Improved |
| add progressive energy-based TTA | pretrained UNet · ACDC to M&M LV | DSC (%): 47.69 -> 61.84 | Improved |
| add progressive energy-based TTA | pretrained UNet · M&M to ACDC LV | DSC (%): 40.84 -> 59.97 | Improved |
| use energy TTA instead of TEA | UNet TTA · ACDC to LVQuant LV | DSC (%): 67.96 -> 76.93 | Improved |
| use energy TTA instead of TEA | MedNeXt TTA · ACDC to LVQuant LV | DSC (%): 75.85 -> 76.22 | Improved |
| use energy TTA instead of TEA | TTA on GMSC · GMSC spinal cord avg | DSC (%): 77.3 -> 80.9 | Improved |
| use energy TTA instead of InTENT | TTA on CHN lung · CHN to MCU/JSRT avg | DSC (%): 95.9 -> 96.2 | Improved |
| add adversarial perturbations for energy-model negatives (ablation) | energy model · ACDC to LVQuant LV | DSC (%): 64.72 -> 76.93 | Improved |
| add spatial augmentation for energy-model negatives (ablation) | energy model · ACDC to LVQuant LV | DSC (%): 76.15 -> 76.93 | Improved |
| increase adaptation iterations from 1 to 10 (ablation) | energy TTA · ACDC to LVQuant LV | DSC (%): 64.85 -> 76.93 | Improved |
| reduce batch size from 4 to 1 (ablation) | energy TTA · GMSC 1 to 3 | DSC (%): 77.7 -> 83.5 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TENT — Tent: Fully Test-Time Adaptation by Entropy Minimization (ICLR 2021) · better on 5 of 5 shared results
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
- Test-time Adaptation for Foundation Medical Segmentation Model Without Parametric Updates (ICCV 2025 · similarity 0.89)
- Gradient Alignment Improves Test-Time Adaptation for Medical Image Segmentation (AAAI 2025 · similarity 0.87)
- Each Test Image Deserves A Specific Prompt: Continual Test-Time Adaptation for 2D Medical Image Segmentation (CVPR 2024 · similarity 0.86)
- Efficient Deformable Convolutional Prompt for Continual Test-Time Adaptation in Medical Image Segmentation (AAAI 2025 · similarity 0.86)
- Training Like a Medical Resident: Context-Prior Learning Toward Universal Medical Image Segmentation (CVPR 2024 · similarity 0.85)
- Knowledge Transfer Scaling Laws for 3D Medical Imaging (NeurIPS 2026 · 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.
