Paper: Contrastive self-supervised learning for spatio-temporal analysis of lung ultrasound videos


Contrastive self-supervised learning for spatio-temporal analysis of lung ultrasound videos

Li Chen*, Jonathan Rubin*†, Jiahong Ouyang, Naveen Balaraju, Shubham Patil, Courosh Mehanian, Sourabh Kulhare, Rachel Millin, Kenton W. Gregory, Cynthia R. Gregory, Meihua Zhu, David O. Kessler, Laurie Malia, Almaz Dessie, Joni Rabiner, Di Coneybeare, Bo Shopsin, Andrew Hersh, Cristian Madar, Jeffrey Shupp, Laura S. Johnson, Jacob Avila, Kristin Dwyer, Peter Weimersheimer, Balasundar Raju, Jochen Kruecker, Alvin Chen.

* equal contribution

work done while employed at Philips

 IEEE International Symposium on Biomedical Imaging (ISBI) , 2023, Cartagena, Colombia (April 18 - 21).


Abstract: 

Self-supervised learning (SSL) methods have shown promise for medical imaging applications by learning meaningful visual representations, even when the amount of labeled data is limited. Here, we extend state-of-the-art contrastive learning SSL methods to 2D+time medical ultrasound video data by introducing a modified encoder and augmentation method capable of learning meaningful spatio-temporal representations, without requiring constraints on the input data. We evaluate our method on the challenging clinical task of identifying lung consolidations (an important pathological feature) in ultrasound videos. Using a multi-center dataset of over 27k lung ultrasound videos acquired from over 500 patients, we show that our method can significantly improve performance on downstream localization and classification of lung consolidation. Comparisons against baseline models trained without SSL show that the proposed methods are particularly advantageous when the size of labeled training data is limited (e.g., as little as 5% of the training set).


Index Terms—Self-supervised learning, contrastive learning, spatio-temporal augmentation, lung ultrasound


This is a joint work from multiple clinical sites. The annotations require huge amount of human efforts. So this paper has  the most authors I ever published. 

isbi.png


Last Article

Comment 评论



Share 分享

New Users 最新加入

  • hokurikustr

  • refrain

New comments 最新评论

test123: aasdas Details Apr 13 16:39
admin: Thanks! Details Apr 09 11:46
admin: Google map api Details Apr 09 11:46
lqj12: cooooooooool Details Apr 08 21:34
Yunhan Huang: 这个功能是如何实现的? Details Apr 08 13:23