Physiology-Aware Masked Cross-Modal Reconstruction for Biosignal Representation Learning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910184811528192 |
|---|---|
| author | Zhou, Hao Lee, Simon A. Tanade, Cyrus Chun, Keum San Lee, Juhyeon Gwak, Migyeong Thukral, Megha Sung, Justin Hwang, Eugene Morshed, Mehrab Bin Zhu, Li Nathan, Viswam Rahman, Md Mahbubur Venkatraman, Subramaniam Desai, Sharanya Arcot |
| author_facet | Zhou, Hao Lee, Simon A. Tanade, Cyrus Chun, Keum San Lee, Juhyeon Gwak, Migyeong Thukral, Megha Sung, Justin Hwang, Eugene Morshed, Mehrab Bin Zhu, Li Nathan, Viswam Rahman, Md Mahbubur Venkatraman, Subramaniam Desai, Sharanya Arcot |
| contents | Biosignals acquired from different locations on the body often provide temporally ordered views of the same underlying physiological process. However, most existing self supervised learning methods treat these signals as interchangeable views, overlooking the directional temporal dynamics that link them. A canonical example is the relationship between electrocardiography (ECG), which captures the electrical activation initiating each heartbeat, and photoplethysmography (PPG), which records the resulting peripheral pulse delayed by vascular dynamics. To capture this structured relationship, we introduce xMAE, a biosignal pretraining framework that leverages masked cross modal reconstruction across temporally ordered biosignals as a training time constraint to encourage physiologically meaningful timing structure in the learned representations. We show that pretraining with xMAE yields representations that outperform both unimodal and multimodal baselines on 15 of 19 downstream tasks, including cardiovascular outcome prediction, abnormal laboratory test detection, sleep staging, and demographic inference, while generalizing across devices, body locations, and acquisition settings. Further analysis suggests that the ECG PPG timing structure is reflected in the learned PPG representations. More broadly, xMAE demonstrates the effectiveness of incorporating temporal structure into multimodal pretraining when signals observe different stages of a shared underlying process. Code is available at https://github.com/hzhou3/xMAE. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_00973 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Physiology-Aware Masked Cross-Modal Reconstruction for Biosignal Representation Learning Zhou, Hao Lee, Simon A. Tanade, Cyrus Chun, Keum San Lee, Juhyeon Gwak, Migyeong Thukral, Megha Sung, Justin Hwang, Eugene Morshed, Mehrab Bin Zhu, Li Nathan, Viswam Rahman, Md Mahbubur Venkatraman, Subramaniam Desai, Sharanya Arcot Machine Learning Artificial Intelligence Signal Processing Biosignals acquired from different locations on the body often provide temporally ordered views of the same underlying physiological process. However, most existing self supervised learning methods treat these signals as interchangeable views, overlooking the directional temporal dynamics that link them. A canonical example is the relationship between electrocardiography (ECG), which captures the electrical activation initiating each heartbeat, and photoplethysmography (PPG), which records the resulting peripheral pulse delayed by vascular dynamics. To capture this structured relationship, we introduce xMAE, a biosignal pretraining framework that leverages masked cross modal reconstruction across temporally ordered biosignals as a training time constraint to encourage physiologically meaningful timing structure in the learned representations. We show that pretraining with xMAE yields representations that outperform both unimodal and multimodal baselines on 15 of 19 downstream tasks, including cardiovascular outcome prediction, abnormal laboratory test detection, sleep staging, and demographic inference, while generalizing across devices, body locations, and acquisition settings. Further analysis suggests that the ECG PPG timing structure is reflected in the learned PPG representations. More broadly, xMAE demonstrates the effectiveness of incorporating temporal structure into multimodal pretraining when signals observe different stages of a shared underlying process. Code is available at https://github.com/hzhou3/xMAE. |
| title | Physiology-Aware Masked Cross-Modal Reconstruction for Biosignal Representation Learning |
| topic | Machine Learning Artificial Intelligence Signal Processing |
| url | https://arxiv.org/abs/2605.00973 |