A robust PPG foundation model using multimodal physiological supervision

Eloy Geenjaar, Vince D. Calhoun, Scott Daly, Gouthaman Kv, Lie Lu, Trisha Mittal, Daniel P. Darcy
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:34418-34441, 2026.

Abstract

Photoplethysmography (PPG), a non-invasive measure of changes in blood volume, is widely used in both wearable devices and clinical settings. Recent PPG foundation models either use open-source ICU datasets with pretraining paradigms that require curated data and thus complicate generalization to field-like data, or use closed-source field-like PPG data. In contrast, we propose a PPG foundation model that does not require high-quality or field-like pretraining data, and instead leverages accompanying electrocardiogram and respiratory signals in ICU datasets to select contrastive samples during pretraining. Our approach allows the model to retain and learn from noisy PPG segments, improving robustness at inference. Our model, pretrained on 3x fewer subjects than existing state-of-the-art approaches, achieves performance improvements on 14 out of 15 diverse downstream tasks, including field-like daily activity and heart rate prediction. Our results demonstrate that multimodal supervision can integrate complementary physiological information to improve the robustness of PPG foundation models and enhance their generalization to consumer-grade data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-geenjaar26a, title = {A robust {PPG} foundation model using multimodal physiological supervision}, author = {Geenjaar, Eloy and Calhoun, Vince D. and Daly, Scott and Kv, Gouthaman and Lu, Lie and Mittal, Trisha and Darcy, Daniel P.}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {34418--34441}, 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/geenjaar26a/geenjaar26a.pdf}, url = {https://proceedings.mlr.press/v306/geenjaar26a.html}, abstract = {Photoplethysmography (PPG), a non-invasive measure of changes in blood volume, is widely used in both wearable devices and clinical settings. Recent PPG foundation models either use open-source ICU datasets with pretraining paradigms that require curated data and thus complicate generalization to field-like data, or use closed-source field-like PPG data. In contrast, we propose a PPG foundation model that does not require high-quality or field-like pretraining data, and instead leverages accompanying electrocardiogram and respiratory signals in ICU datasets to select contrastive samples during pretraining. Our approach allows the model to retain and learn from noisy PPG segments, improving robustness at inference. Our model, pretrained on 3x fewer subjects than existing state-of-the-art approaches, achieves performance improvements on 14 out of 15 diverse downstream tasks, including field-like daily activity and heart rate prediction. Our results demonstrate that multimodal supervision can integrate complementary physiological information to improve the robustness of PPG foundation models and enhance their generalization to consumer-grade data.} }
Endnote
%0 Conference Paper %T A robust PPG foundation model using multimodal physiological supervision %A Eloy Geenjaar %A Vince D. Calhoun %A Scott Daly %A Gouthaman Kv %A Lie Lu %A Trisha Mittal %A Daniel P. Darcy %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-geenjaar26a %I PMLR %P 34418--34441 %U https://proceedings.mlr.press/v306/geenjaar26a.html %V 306 %X Photoplethysmography (PPG), a non-invasive measure of changes in blood volume, is widely used in both wearable devices and clinical settings. Recent PPG foundation models either use open-source ICU datasets with pretraining paradigms that require curated data and thus complicate generalization to field-like data, or use closed-source field-like PPG data. In contrast, we propose a PPG foundation model that does not require high-quality or field-like pretraining data, and instead leverages accompanying electrocardiogram and respiratory signals in ICU datasets to select contrastive samples during pretraining. Our approach allows the model to retain and learn from noisy PPG segments, improving robustness at inference. Our model, pretrained on 3x fewer subjects than existing state-of-the-art approaches, achieves performance improvements on 14 out of 15 diverse downstream tasks, including field-like daily activity and heart rate prediction. Our results demonstrate that multimodal supervision can integrate complementary physiological information to improve the robustness of PPG foundation models and enhance their generalization to consumer-grade data.
APA
Geenjaar, E., Calhoun, V.D., Daly, S., Kv, G., Lu, L., Mittal, T. & Darcy, D.P.. (2026). A robust PPG foundation model using multimodal physiological supervision. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:34418-34441 Available from https://proceedings.mlr.press/v306/geenjaar26a.html.

Related Material