mlsanityData to 7 Oct 2026

Progressive Test Time Energy Adaptation for Medical Image Segmentation

arXiv 2503.16616 ↗First posted 20 Mar 2025cs.CVcited by 5

Not on the map. Why?

Abstract (arXiv)

We propose a model-agnostic, progressive test-time energy adaptation approach for medical image segmentation. Maintaining model performance across diverse medical datasets is challenging, as distribution shifts arise from inconsistent imaging protocols and patient variations. Unlike domain adaptation methods that require multiple passes through target data - impractical in clinical settings - our approach adapts pretrained models progressively as they process test data. Our method leverages a shape energy model trained on source data, which assigns an energy score at the patch level to segmentation maps: low energy represents in-distribution (accurate) shapes, while high energy signals out-of-distribution (erroneous) predictions. By minimizing this energy score at test time, we refine the segmentation model to align with the target distribution. To validate the effectiveness and adaptability, we evaluated our framework on eight public MRI (bSSFP, T1- and T2-weighted) and X-ray datasets spanning cardiac, spinal cord, and lung segmentation. We consistently outperform baselines both quantitatively and qualitatively.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 42 of its 72 references that are in the map. First posted on arXiv 20 Mar 2025.

Builds on

42 of 72 references.

References · 42

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  4. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  5. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  6. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  7. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
  8. 2014Unsupervised Domain Adaptation by Backpropagationcited by 1,906
  9. 2018Moment Matching for Multi-Source Domain Adaptationcited by 1,327
  10. 2022Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Imagescited by 823
  11. 2022Continual Test-Time Domain Adaptationcited by 520
  12. 2022Efficient Test-Time Model Adaptation without Forgettingcited by 467
30 more
  1. 2020Improving robustness against common corruptions by covariate shift adaptationcited by 427
  2. 2024U-Mamba: Enhancing Long-range Dependency for Biomedical Image Segmentationcited by 412
  3. 2019Your Classifier is Secretly an Energy Based Model and You Should Treat it Like Onecited by 402
  4. 2021MEMO: Test Time Robustness via Adaptation and Augmentationcited by 376
  5. 2018A Survey of Unsupervised Deep Domain Adaptationcited by 361
  6. 2021Domain Adaptation for Medical Image Analysis: A Surveycited by 351
  7. 2020A Brief Review of Domain Adaptationcited by 287
  8. 2017PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examplescited by 243
  9. 2019Deep Learning for Segmentation using an Open Large-Scale Dataset in 2D Echocardiographycited by 243
  10. 2024nnU-Net Revisited: A Call for Rigorous Validation in 3D Medical Image Segmentationcited by 235
  11. 2021Source-Free Domain Adaptation for Semantic Segmentationcited by 168
  12. 2023MedNeXt: Transformer-driven Scaling of ConvNets for Medical Image Segmentationcited by 161
  13. 2024Mamba-UNet: UNet-Like Pure Visual Mamba for Medical Image Segmentationcited by 133
  14. 2023EcoTTA: Memory-Efficient Continual Test-time Adaptation via Self-distilled Regularizationcited by 123
  15. 2023A Comprehensive Survey on Source-free Domain Adaptationcited by 104
  16. 2018Unsupervised domain adaptation for medical imaging segmentation with self-ensemblingcited by 94
  17. 2020Domain Adaptation without Source Datacited by 88
  18. 2021MT3: Meta Test-Time Training for Self-Supervised Test-Time Adaptioncited by 65
  19. 2019Test-Time Training with Self-Supervision for Generalization under Distribution Shiftscited by 55
  20. 2020Post-DAE: Anatomically Plausible Segmentation via Post-Processing with Denoising Autoencoderscited by 38
  21. 2018Divergence Triangle for Joint Training of Generator Model, Energy-based Model, and Inference Modelcited by 34
  22. 2019Unsupervised Domain Adaptation via Regularized Conditional Alignmentcited by 34
  23. 2022On-the-Fly Test-time Adaptation for Medical Image Segmentationcited by 34
  24. 2025Model Adaptation: Unsupervised Domain Adaptation without Source Datacited by 34
  25. 2022MyoPS: A Benchmark of Myocardial Pathology Segmentation Combining Three-Sequence Cardiac Magnetic Resonance Imagescited by 29
  26. 2024Test-Time Adaptation for Depth Completioncited by 26
  27. 2023TEA: Test-time Energy Adaptationcited by 25
  28. 2019Implicit Generation and Generalization in Energy-Based Modelscited by 21
  29. 2023From Denoising Training to Test-Time Adaptation: Enhancing Domain Generalization for Medical Image Segmentationcited by 20
  30. 2020Fully Automated Left Atrium Segmentation from Anatomical Cine Long-axis MRI Sequences using Deep Convolutional Neural Network with Unscented Kalman Filtercited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 2; 2026: 3.

5 papers.

  1. 2025Extending Foundational Monocular Depth Estimators to Fisheye Cameras with Calibration Tokenscited by 9
  2. 2025ETA: Energy-based Test-time Adaptation for Depth Completioncited by 4
  3. 2026VisTa3D: A Dataset and Benchmark for Thin Object Reconstruction from Vision, Tactile, and 3D Point Cloudscited by 1
  4. 2026Test-Time Adaptation for Anomaly Segmentation via Topology-Aware Optimal Transport Chainingcited by 0
  5. 2026From Perspective to Fisheye Depth Estimation and Open-Vocabulary Segmentationcited by 0