Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
97 tokens/sec
GPT-4o
53 tokens/sec
Gemini 2.5 Pro Pro
44 tokens/sec
o3 Pro
5 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Medical Image Segmentation with InTEnt: Integrated Entropy Weighting for Single Image Test-Time Adaptation (2402.09604v2)

Published 14 Feb 2024 in cs.CV and cs.AI

Abstract: Test-time adaptation (TTA) refers to adapting a trained model to a new domain during testing. Existing TTA techniques rely on having multiple test images from the same domain, yet this may be impractical in real-world applications such as medical imaging, where data acquisition is expensive and imaging conditions vary frequently. Here, we approach such a task, of adapting a medical image segmentation model with only a single unlabeled test image. Most TTA approaches, which directly minimize the entropy of predictions, fail to improve performance significantly in this setting, in which we also observe the choice of batch normalization (BN) layer statistics to be a highly important yet unstable factor due to only having a single test domain example. To overcome this, we propose to instead integrate over predictions made with various estimates of target domain statistics between the training and test statistics, weighted based on their entropy statistics. Our method, validated on 24 source/target domain splits across 3 medical image datasets surpasses the leading method by 2.9% Dice coefficient on average.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (38)
  1. Integrated likelihood methods for eliminating nuisance parameters. Statistical science, pp.  1–22, 1999.
  2. Sharpness-aware minimization for efficiently improving generalization. In International Conference on Learning Representations, 2020.
  3. An ensemble classification-based approach applied to retinal blood vessel segmentation. IEEE Transactions on Biomedical Engineering, 59(9):2538–2548, 2012.
  4. Decorate the newcomers: Visual domain prompt for continual test time adaptation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, pp.  7595–7603, 2023.
  5. Back to the source: Diffusion-driven adaptation to test-time corruption. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  11786–11796, 2023.
  6. Domain adaptation for medical image analysis: a survey. IEEE Transactions on Biomedical Engineering, 69(3):1173–1185, 2021.
  7. On calibration of modern neural networks. In International conference on machine learning, pp.  1321–1330. PMLR, 2017.
  8. The real-world-weight cross-entropy loss function: Modeling the costs of mislabeling. IEEE access, 8:4806–4813, 2019.
  9. Bayesian model averaging: a tutorial (with comments by m. clyde, david draper and ei george, and a rejoinder by the authors. Statistical science, 14(4):382–417, 1999.
  10. Fully test-time adaptation for image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention, 2021.
  11. Automated separation of binary overlapping trees in low-contrast color retinal images. In Medical Image Computing and Computer-Assisted Intervention–MICCAI 2013: 16th International Conference, Nagoya, Japan, September 22-26, 2013, Proceedings, Part II 16, pp.  436–443. Springer, 2013.
  12. Batch normalization: Accelerating deep network training by reducing internal covariate shift. In International Conference on Machine Learning, 2015.
  13. Two public chest x-ray datasets for computer-aided screening of pulmonary diseases. Quantitative imaging in medicine and surgery, 4 6:475–7, 2014.
  14. Test-time adaptable neural networks for robust medical image segmentation. Medical Image Analysis, 68:101907, 2021.
  15. Sita: Single image test-time adaptation. ArXiv, abs/2112.02355, 2021.
  16. Adam: A method for stochastic optimization. In Bengio, Y. and LeCun, Y. (eds.), 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings, 2015. URL http://arxiv.org/abs/1412.6980.
  17. Wilds: A benchmark of in-the-wild distribution shifts. ArXiv, abs/2012.07421, 2020. URL https://api.semanticscholar.org/CorpusID:229156320.
  18. Reverse engineering breast mris: Predicting acquisition parameters directly from images. In Medical Imaging with Deep Learning, 2023.
  19. Domain generalization for medical imaging classification with linear-dependency regularization. Advances in neural information processing systems, 33:3118–3129, 2020.
  20. A comprehensive survey on test-time adaptation under distribution shifts. arXiv preprint arXiv:2303.15361, 2023.
  21. Applications of deep learning to mri images: A survey. Big Data Mining and Analytics, 1(1):1–18, 2018.
  22. Single-domain generalization in medical image segmentation via test-time adaptation from shape dictionary. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, pp.  1756–1764, 2022.
  23. Adapting off-the-shelf source segmenter for target medical image segmentation. In Medical Image Computing and Computer Assisted Intervention – MICCAI 2021, pp.  549–559, 2021.
  24. Test-time adaptation with calibration of medical image classification nets for label distribution shift. In International Conference on Medical Image Computing and Computer-Assisted Intervention, 2022.
  25. Improved denoising diffusion probabilistic models. In International Conference on Machine Learning, pp.  8162–8171. PMLR, 2021.
  26. Towards stable test-time adaptation in dynamic wild world. In The Eleventh International Conference on Learning Representations, 2023.
  27. Improvement of vessel segmentation by matched filtering in colour retinal images. In World Congress on Medical Physics and Biomedical Engineering, September 7-12, 2009, Munich, Germany: Vol. 25/11 Biomedical Engineering for Audiology, Ophthalmology, Emergency & Dental Medicine, pp.  327–330. Springer, 2009.
  28. Spinal cord grey matter segmentation challenge. Neuroimage, 152:312 – 329, 2017.
  29. Dataset shift in machine learning. 2009.
  30. Improving robustness against common corruptions by covariate shift adaptation. Advances in neural information processing systems, 33:11539–11551, 2020.
  31. Development of a digital image database for chest radiographs with and without a lung nodule: receiver operating characteristic analysis of radiologists’ detection of pulmonary nodules. AJR. American journal of roentgenology, 174 1:71–4, 2000.
  32. Test-time training with self-supervision for generalization under distribution shifts. In International Conference on Machine Learning, 2019.
  33. On-the-fly test-time adaptation for medical image segmentation. In Medical Imaging with Deep Learning, 2023. URL https://openreview.net/forum?id=UQDalTzrEg.
  34. Advent: Adversarial entropy minimization for domain adaptation in semantic segmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  2517–2526, 2019.
  35. Tent: Fully test-time adaptation by entropy minimization. In International Conference on Learning Representations, 2021.
  36. Continual test-time domain adaptation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  7201–7211, 2022.
  37. Memo: Test time robustness via adaptation and augmentation. Advances in Neural Information Processing Systems, 35:38629–38642, 2022.
  38. On pitfalls of test-time adaptation. In ICLR 2023 Workshop on Pitfalls of limited data and computation for Trustworthy ML, 2023. URL https://openreview.net/forum?id=0Go_RsG_dYn.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (4)
  1. Haoyu Dong (55 papers)
  2. Nicholas Konz (22 papers)
  3. Hanxue Gu (22 papers)
  4. Maciej A. Mazurowski (51 papers)

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets