PyHealth 2.0: A Comprehensive Open-Source Toolkit for Accessible and Reproducible Clinical Deep Learning

John Wu, Yongda Fan, Zhenbang Wu, Paul Landes, Eric Schrock, Sayeed Sajjad Razin, Arjun Chatterjee, Naveen Baskaran, Joshua Steier, Andrea Fitzpatrick, Bilal Arif, Rian Atri, Jathurshan Pradeepkumar, Siddhartha Laghuvarapu, Junyi Gao, Adam Richard Cross, Jimeng Sun
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:135821-135859, 2026.

Abstract

Difficulty replicating baselines, high computational costs, and required domain expertise create persistent barriers to clinical AI research. To address these challenges, we introduce PyHealth 2.0, an enhanced clinical deep learning toolkit that enables predictive modeling in as few as 7 lines of code. PyHealth 2.0 offers three key contributions: (1) a comprehensive toolkit addressing reproducibility and compatibility challenges by unifying 15+ datasets, 20+ clinical tasks, 25+ models, 5+ interpretability methods, and uncertainty quantification including conformal prediction within a single framework that supports diverse clinical data modalities—signals, imaging, and electronic health records—with translation of 5+ medical coding standards; (2) accessibility-focused design accommodating multimodal data and diverse computational resources with up to 39$\times$ faster processing and 20$\times$ lower memory usage, enabling work from 16GB laptops to production systems; and (3) an active open-source community of 400+ members lowering domain expertise barriers through extensive documentation, reproducible research contributions, and collaborations with academic health systems and industry partners, including multi-language support via RHealth. PyHealth 2.0 establishes an open-source foundation and community advancing accessible, reproducible healthcare AI.

Cite this Paper


BibTeX
@InProceedings{pmlr-v306-wu26n, title = {{P}y{H}ealth 2.0: A Comprehensive Open-Source Toolkit for Accessible and Reproducible Clinical Deep Learning}, author = {Wu, John and Fan, Yongda and Wu, Zhenbang and Landes, Paul and Schrock, Eric and Razin, Sayeed Sajjad and Chatterjee, Arjun and Baskaran, Naveen and Steier, Joshua and Fitzpatrick, Andrea and Arif, Bilal and Atri, Rian and Pradeepkumar, Jathurshan and Laghuvarapu, Siddhartha and Gao, Junyi and Cross, Adam Richard and Sun, Jimeng}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {135821--135859}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/wu26n/wu26n.pdf}, url = {https://proceedings.mlr.press/v306/wu26n.html}, abstract = {Difficulty replicating baselines, high computational costs, and required domain expertise create persistent barriers to clinical AI research. To address these challenges, we introduce PyHealth 2.0, an enhanced clinical deep learning toolkit that enables predictive modeling in as few as 7 lines of code. PyHealth 2.0 offers three key contributions: (1) a comprehensive toolkit addressing reproducibility and compatibility challenges by unifying 15+ datasets, 20+ clinical tasks, 25+ models, 5+ interpretability methods, and uncertainty quantification including conformal prediction within a single framework that supports diverse clinical data modalities—signals, imaging, and electronic health records—with translation of 5+ medical coding standards; (2) accessibility-focused design accommodating multimodal data and diverse computational resources with up to 39$\times$ faster processing and 20$\times$ lower memory usage, enabling work from 16GB laptops to production systems; and (3) an active open-source community of 400+ members lowering domain expertise barriers through extensive documentation, reproducible research contributions, and collaborations with academic health systems and industry partners, including multi-language support via RHealth. PyHealth 2.0 establishes an open-source foundation and community advancing accessible, reproducible healthcare AI.} }
Endnote
%0 Conference Paper %T PyHealth 2.0: A Comprehensive Open-Source Toolkit for Accessible and Reproducible Clinical Deep Learning %A John Wu %A Yongda Fan %A Zhenbang Wu %A Paul Landes %A Eric Schrock %A Sayeed Sajjad Razin %A Arjun Chatterjee %A Naveen Baskaran %A Joshua Steier %A Andrea Fitzpatrick %A Bilal Arif %A Rian Atri %A Jathurshan Pradeepkumar %A Siddhartha Laghuvarapu %A Junyi Gao %A Adam Richard Cross %A Jimeng Sun %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-wu26n %I PMLR %P 135821--135859 %U https://proceedings.mlr.press/v306/wu26n.html %V 306 %X Difficulty replicating baselines, high computational costs, and required domain expertise create persistent barriers to clinical AI research. To address these challenges, we introduce PyHealth 2.0, an enhanced clinical deep learning toolkit that enables predictive modeling in as few as 7 lines of code. PyHealth 2.0 offers three key contributions: (1) a comprehensive toolkit addressing reproducibility and compatibility challenges by unifying 15+ datasets, 20+ clinical tasks, 25+ models, 5+ interpretability methods, and uncertainty quantification including conformal prediction within a single framework that supports diverse clinical data modalities—signals, imaging, and electronic health records—with translation of 5+ medical coding standards; (2) accessibility-focused design accommodating multimodal data and diverse computational resources with up to 39$\times$ faster processing and 20$\times$ lower memory usage, enabling work from 16GB laptops to production systems; and (3) an active open-source community of 400+ members lowering domain expertise barriers through extensive documentation, reproducible research contributions, and collaborations with academic health systems and industry partners, including multi-language support via RHealth. PyHealth 2.0 establishes an open-source foundation and community advancing accessible, reproducible healthcare AI.
APA
Wu, J., Fan, Y., Wu, Z., Landes, P., Schrock, E., Razin, S.S., Chatterjee, A., Baskaran, N., Steier, J., Fitzpatrick, A., Arif, B., Atri, R., Pradeepkumar, J., Laghuvarapu, S., Gao, J., Cross, A.R. & Sun, J.. (2026). PyHealth 2.0: A Comprehensive Open-Source Toolkit for Accessible and Reproducible Clinical Deep Learning. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:135821-135859 Available from https://proceedings.mlr.press/v306/wu26n.html.

Related Material