Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
175 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Intra-video Positive Pairs in Self-Supervised Learning for Ultrasound (2403.07715v1)

Published 12 Mar 2024 in eess.IV and cs.CV

Abstract: Self-supervised learning (SSL) is one strategy for addressing the paucity of labelled data in medical imaging by learning representations from unlabelled images. Contrastive and non-contrastive SSL methods produce learned representations that are similar for pairs of related images. Such pairs are commonly constructed by randomly distorting the same image twice. The videographic nature of ultrasound offers flexibility for defining the similarity relationship between pairs of images. In this study, we investigated the effect of utilizing proximal, distinct images from the same B-mode ultrasound video as pairs for SSL. Additionally, we introduced a sample weighting scheme that increases the weight of closer image pairs and demonstrated how it can be integrated into SSL objectives. Named Intra-Video Positive Pairs (IVPP), the method surpassed previous ultrasound-specific contrastive learning methods' average test accuracy on COVID-19 classification with the POCUS dataset by $\ge 1.3\%$. Detailed investigations of IVPP's hyperparameters revealed that some combinations of IVPP hyperparameters can lead to improved or worsened performance, depending on the downstream task. Guidelines for practitioners were synthesized based on the results, such as the merit of IVPP with task-specific hyperparameters, and the improved performance of contrastive methods for ultrasound compared to non-contrastive counterparts.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (48)
  1. Ultrasonography in the emergency department. Critical Care, 20:1–8, 2016.
  2. Point-of-care ultrasound for critically-ill patients: A mini-review of key diagnostic features and protocols. World Journal of Critical Care Medicine, 11(2):70, 2022.
  3. Point-of-Care Ultrasound. Elsevier, Philadelphia, second edition, 2020.
  4. Ultrasound for breast cancer detection globally: a systematic review and meta-analysis. Journal of global oncology, 5:1–17, 2019.
  5. Ultrasound in sports medicine: relevance of emerging techniques to clinical care of athletes. Sports medicine, 42:665–680, 2012.
  6. A comparison of the accuracy of ultrasound and computed tomography in common diagnoses causing acute abdominal pain. European radiology, 21:1535–1545, 2011.
  7. Test characteristics of ultrasonography for the detection of pneumothorax: a systematic review and meta-analysis. Chest, 141(3):703–708, 2012.
  8. Accuracy of lung ultrasound for the diagnosis of consolidations when compared to chest computed tomography. The American journal of emergency medicine, 33(5):620–625, 2015.
  9. Use of point-of-care ultrasound in the emergency department: insights from the 2012 medicare national payment data set. Journal of Ultrasound in Medicine, 35(11):2467–2474, 2016.
  10. Effect of interventional program on the utilization of pacs in point-of-care ultrasound. Journal of digital imaging, 29:701–705, 2016.
  11. USCL: Pretraining Deep Ultrasound Image Diagnosis Model Through Video Contrastive Representation Learning. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th International Conference, Strasbourg, France, September 27–October 1, 2021, Proceedings, Part VIII 24, pages 627–637. Springer, 2021.
  12. Unsupervised Contrastive Learning of Image Representations from Ultrasound Videos with Hard Negative Mining. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 13434 LNCS:423–433, 2022. ISBN: 9783031164392 Publisher: Springer Science and Business Media Deutschland GmbH.
  13. Contrastive and non-contrastive self-supervised learning recover global and local spectral embedding methods. In S. Koyejo, S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, and A. Oh, editors, Advances in Neural Information Processing Systems, volume 35, pages 26671–26685. Curran Associates, Inc., 2022.
  14. A simple framework for contrastive learning of visual representations. In International conference on machine learning, pages 1597–1607. PMLR, 2020.
  15. Bootstrap your own latent-a new approach to self-supervised learning. Advances in neural information processing systems, 33:21271–21284, 2020.
  16. Barlow twins: Self-supervised learning via redundancy reduction. In International Conference on Machine Learning, pages 12310–12320, 2021.
  17. VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning. In International Conference on Learning Representations, 2022.
  18. Deep metric learning via lifted structured feature embedding. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4004–4012, 2016.
  19. Momentum contrast for unsupervised visual representation learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 9729–9738, 2020.
  20. With a little help from my friends: Nearest-neighbor contrastive learning of visual representations. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 9588–9597, 2021.
  21. Decoupled contrastive learning. In European Conference on Computer Vision, pages 668–684. Springer, 2022.
  22. On the duality between contrastive and non-contrastive self-supervised learning. arXiv preprint arXiv:2206.02574, 2022.
  23. mixup: Beyond empirical risk minimization. arXiv preprint arXiv:1710.09412, 2017.
  24. POCOVID-Net: Automatic Detection of COVID-19 From a New Lung Ultrasound Imaging Dataset (POCUS). arXiv preprint arXiv:2004.12084, 2020.
  25. Generating and weighting semantically consistent sample pairs for ultrasound contrastive learning. IEEE Transactions on Medical Imaging, 2022.
  26. Hico: Hierarchical contrastive learning for ultrasound video model pretraining. In Proceedings of the Asian Conference on Computer Vision, pages 229–246, 2022.
  27. Emerging properties in self-supervised vision transformers. In Proceedings of the IEEE/CVF international conference on computer vision, pages 9650–9660, 2021.
  28. Benchmarking self-supervised representation learning from a million cardiac ultrasound images. In 2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), pages 529–532. IEEE, 2022.
  29. Semi-supervised anatomy tracking with contrastive representation learning in ultrasound sequences. In 2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI), pages 1–5. IEEE, 2023.
  30. Exploring the utility of self-supervised pretraining strategies for the detection of absent lung sliding in m-mode lung ultrasound. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 3076–3085, 2023.
  31. Self-supervised pretraining improves performance and inference efficiency in multiple lung ultrasound interpretation tasks. arXiv preprint arXiv:2309.02596, 2023.
  32. Accurate assessment of the lung sliding artefact on lung ultrasonography using a deep learning approach. Computers in Biology and Medicine, 148:105953, 2022.
  33. Detecting the absence of lung sliding in lung ultrasounds using deep learning. Applied Sciences, 11(15):6976, 2021.
  34. Butterfly Network. Covid-19 ultrasound gallery. https://www.butterflynetwork.com/covid19/covid-19-ultrasound-gallery, 2020. Accessed: September 20, 2020.
  35. Automation of Lung Ultrasound Interpretation via Deep Learning for the Classification of Normal Versus Abnormal Lung Parenchyma: A Multicenter Study. Diagnostics, 11(11):2049, 2021.
  36. Enhancing annotation efficiency with machine learning: Automated partitioning of a lung ultrasound dataset by view. Diagnostics, 12(10):2351, 2022.
  37. A bedside ultrasound sign ruling out pneumothorax in the critically iii: lung sliding. Chest, 108(5):1345–1348, 1995.
  38. Ultrasound diagnosis of occult pneumothorax. Critical care medicine, 33(6):1231–1238, 2005.
  39. Daniel A Lichtenstein. Whole body ultrasonography in the critically ill. Springer Science & Business Media, 2010.
  40. Big Self-Supervised Models Advance Medical Image Classification. In Proceedings of the IEEE International Conference on Computer Vision, pages 3458–3468. Institute of Electrical and Electronics Engineers Inc., 2021. ISSN: 15505499.
  41. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016.
  42. Searching for MobileNetV3. In Proceedings of the IEEE/CVF international conference on computer vision, pages 1314–1324, 2019.
  43. Large batch optimization for deep learning: Training bert in 76 minutes. arXiv preprint arXiv:1904.00962, 2019.
  44. Randall Balestriero. Neural decision trees. arXiv preprint arXiv:1702.07360, 2017.
  45. Rethinking the value of labels for improving class-imbalanced learning. Advances in neural information processing systems, 33:19290–19301, 2020.
  46. Self-supervised learning is more robust to dataset imbalance. arXiv preprint arXiv:2110.05025, 2021.
  47. Dive into the details of self-supervised learning for medical image analysis. Medical Image Analysis, 89:102879, 2023.
  48. Self-supervised video pretraining yields robust and more human-aligned visual representations. In Thirty-seventh Conference on Neural Information Processing Systems, 2023.

Summary

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

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