A comparison of self-supervised pretraining approaches for predicting disease risk from chest radiograph images

Yanru Chen, Michael T Lu, Vineet K Raghu
Medical Imaging with Deep Learning, PMLR 227:1826-1858, 2024.

Abstract

Deep learning is the state-of-the-art for medical imaging tasks, but requires large, labeled datasets. For risk prediction, large datasets are rare since they require both imaging and follow-up (e.g., diagnosis codes). However, the release of publicly available imaging data with diagnostic labels presents an opportunity for self and semi-supervised approaches to improve label efficiency for risk prediction. Though several studies have compared self-supervised approaches in natural image classification, object detection, and medical image interpretation, there is limited data on which approaches learn robust representations for risk prediction. We present a comparison of semi- and self-supervised learning to predict mortality risk using chest x-ray images. We find that a semi-supervised autoencoder outperforms contrastive and transfer learning in internal and external validation.

Cite this Paper

BibTeX
@InProceedings{pmlr-v227-chen24c, title = {A comparison of self-supervised pretraining approaches for predicting disease risk from chest radiograph images}, author = {Chen, Yanru and Lu, Michael T and Raghu, Vineet K}, booktitle = {Medical Imaging with Deep Learning}, pages = {1826--1858}, year = {2024}, editor = {Oguz, Ipek and Noble, Jack and Li, Xiaoxiao and Styner, Martin and Baumgartner, Christian and Rusu, Mirabela and Heinmann, Tobias and Kontos, Despina and Landman, Bennett and Dawant, Benoit}, volume = {227}, series = {Proceedings of Machine Learning Research}, month = {10--12 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v227/chen24c/chen24c.pdf}, url = {https://proceedings.mlr.press/v227/chen24c.html}, abstract = {Deep learning is the state-of-the-art for medical imaging tasks, but requires large, labeled datasets. For risk prediction, large datasets are rare since they require both imaging and follow-up (e.g., diagnosis codes). However, the release of publicly available imaging data with diagnostic labels presents an opportunity for self and semi-supervised approaches to improve label efficiency for risk prediction. Though several studies have compared self-supervised approaches in natural image classification, object detection, and medical image interpretation, there is limited data on which approaches learn robust representations for risk prediction. We present a comparison of semi- and self-supervised learning to predict mortality risk using chest x-ray images. We find that a semi-supervised autoencoder outperforms contrastive and transfer learning in internal and external validation.} }
Endnote
%0 Conference Paper %T A comparison of self-supervised pretraining approaches for predicting disease risk from chest radiograph images %A Yanru Chen %A Michael T Lu %A Vineet K Raghu %B Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2024 %E Ipek Oguz %E Jack Noble %E Xiaoxiao Li %E Martin Styner %E Christian Baumgartner %E Mirabela Rusu %E Tobias Heinmann %E Despina Kontos %E Bennett Landman %E Benoit Dawant %F pmlr-v227-chen24c %I PMLR %P 1826--1858 %U https://proceedings.mlr.press/v227/chen24c.html %V 227 %X Deep learning is the state-of-the-art for medical imaging tasks, but requires large, labeled datasets. For risk prediction, large datasets are rare since they require both imaging and follow-up (e.g., diagnosis codes). However, the release of publicly available imaging data with diagnostic labels presents an opportunity for self and semi-supervised approaches to improve label efficiency for risk prediction. Though several studies have compared self-supervised approaches in natural image classification, object detection, and medical image interpretation, there is limited data on which approaches learn robust representations for risk prediction. We present a comparison of semi- and self-supervised learning to predict mortality risk using chest x-ray images. We find that a semi-supervised autoencoder outperforms contrastive and transfer learning in internal and external validation.
APA
Chen, Y., Lu, M.T. & Raghu, V.K.. (2024). A comparison of self-supervised pretraining approaches for predicting disease risk from chest radiograph images. Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 227:1826-1858 Available from https://proceedings.mlr.press/v227/chen24c.html.

Related Material