CheXphoto: 10,000+ Photos and Transformations of Chest X-rays for Benchmarking Deep Learning Robustness

Nick A. Phillips, Pranav Rajpurkar, Mark Sabini, Rayan Krishnan, Sharon Zhou, Anuj Pareek, Nguyet Minh Phu, Chris Wang, Mudit Jain, Nguyen Duong Du, Steven QH Truong, Andrew Y. Ng, Matthew P. Lungren
Proceedings of the Machine Learning for Health NeurIPS Workshop, PMLR 136:318-327, 2020.

Abstract

Clinical deployment of deep learning algorithms for chest x-ray interpretation requires a solution that can integrate into the vast spectrum of clinical workflows across the world. An appealing approach to scaled deployment is to leverage the ubiquity of smartphones by capturing photos of x-rays to share with clinicians using messaging services like WhatsApp. However, the application of chest x-ray algorithms to photos of chest x-rays requires reliable classification in the presence of artifacts not typically encountered in digital x-rays used to train machine learning models. We introduce CheXphoto, a dataset of smartphone photos and synthetic photographic transformations of chest x-rays sampled from the CheXpert dataset. To generate CheXphoto we (1) automatically and manually captured photos of digital x-rays under different settings, and (2) generated synthetic transformations of digital x-rays targeted to make them look like photos of digital x-rays and x-ray films. We release this dataset as a resource for testing and improving the robustness of deep learning algorithms for automated chest x-ray interpretation on smartphone photos of chest x-rays.

Cite this Paper


BibTeX
@InProceedings{pmlr-v136-phillips20a, title = {CheXphoto: 10,000+ Photos and Transformations of Chest X-rays for Benchmarking Deep Learning Robustness}, author = {Phillips, Nick A. and Rajpurkar, Pranav and Sabini, Mark and Krishnan, Rayan and Zhou, Sharon and Pareek, Anuj and Phu, Nguyet Minh and Wang, Chris and Jain, Mudit and Du, Nguyen Duong and Truong, Steven QH and Ng, Andrew Y. and Lungren, Matthew P.}, booktitle = {Proceedings of the Machine Learning for Health NeurIPS Workshop}, pages = {318--327}, year = {2020}, editor = {Alsentzer, Emily and McDermott, Matthew B. A. and Falck, Fabian and Sarkar, Suproteem K. and Roy, Subhrajit and Hyland, Stephanie L.}, volume = {136}, series = {Proceedings of Machine Learning Research}, month = {11 Dec}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v136/phillips20a/phillips20a.pdf}, url = {https://proceedings.mlr.press/v136/phillips20a.html}, abstract = {Clinical deployment of deep learning algorithms for chest x-ray interpretation requires a solution that can integrate into the vast spectrum of clinical workflows across the world. An appealing approach to scaled deployment is to leverage the ubiquity of smartphones by capturing photos of x-rays to share with clinicians using messaging services like WhatsApp. However, the application of chest x-ray algorithms to photos of chest x-rays requires reliable classification in the presence of artifacts not typically encountered in digital x-rays used to train machine learning models. We introduce CheXphoto, a dataset of smartphone photos and synthetic photographic transformations of chest x-rays sampled from the CheXpert dataset. To generate CheXphoto we (1) automatically and manually captured photos of digital x-rays under different settings, and (2) generated synthetic transformations of digital x-rays targeted to make them look like photos of digital x-rays and x-ray films. We release this dataset as a resource for testing and improving the robustness of deep learning algorithms for automated chest x-ray interpretation on smartphone photos of chest x-rays.} }
Endnote
%0 Conference Paper %T CheXphoto: 10,000+ Photos and Transformations of Chest X-rays for Benchmarking Deep Learning Robustness %A Nick A. Phillips %A Pranav Rajpurkar %A Mark Sabini %A Rayan Krishnan %A Sharon Zhou %A Anuj Pareek %A Nguyet Minh Phu %A Chris Wang %A Mudit Jain %A Nguyen Duong Du %A Steven QH Truong %A Andrew Y. Ng %A Matthew P. Lungren %B Proceedings of the Machine Learning for Health NeurIPS Workshop %C Proceedings of Machine Learning Research %D 2020 %E Emily Alsentzer %E Matthew B. A. McDermott %E Fabian Falck %E Suproteem K. Sarkar %E Subhrajit Roy %E Stephanie L. Hyland %F pmlr-v136-phillips20a %I PMLR %P 318--327 %U https://proceedings.mlr.press/v136/phillips20a.html %V 136 %X Clinical deployment of deep learning algorithms for chest x-ray interpretation requires a solution that can integrate into the vast spectrum of clinical workflows across the world. An appealing approach to scaled deployment is to leverage the ubiquity of smartphones by capturing photos of x-rays to share with clinicians using messaging services like WhatsApp. However, the application of chest x-ray algorithms to photos of chest x-rays requires reliable classification in the presence of artifacts not typically encountered in digital x-rays used to train machine learning models. We introduce CheXphoto, a dataset of smartphone photos and synthetic photographic transformations of chest x-rays sampled from the CheXpert dataset. To generate CheXphoto we (1) automatically and manually captured photos of digital x-rays under different settings, and (2) generated synthetic transformations of digital x-rays targeted to make them look like photos of digital x-rays and x-ray films. We release this dataset as a resource for testing and improving the robustness of deep learning algorithms for automated chest x-ray interpretation on smartphone photos of chest x-rays.
APA
Phillips, N.A., Rajpurkar, P., Sabini, M., Krishnan, R., Zhou, S., Pareek, A., Phu, N.M., Wang, C., Jain, M., Du, N.D., Truong, S.Q., Ng, A.Y. & Lungren, M.P.. (2020). CheXphoto: 10,000+ Photos and Transformations of Chest X-rays for Benchmarking Deep Learning Robustness. Proceedings of the Machine Learning for Health NeurIPS Workshop, in Proceedings of Machine Learning Research 136:318-327 Available from https://proceedings.mlr.press/v136/phillips20a.html.

Related Material