Sharanya Arcot Desai
Sharanya Arcot Desai has 1 paper in the ChatPPG research atlas (2026), linked to 1 model, with work using 4 datasets and 14 co-authors in the atlas. Main topics: ECG, PPG, cross-modal.
Papers
- xMAE: Physiology-Aware Masked Cross-Modal Reconstruction for Biosignal Representation Learning2026 · arXiv preprint
xMAE pretrains on biosignals by reconstructing one signal from another, such as ECG and the delayed PPG pulse, so the model learns physiological timing. It beat unimodal and multimodal baselines on 15 of 19 tasks, from cardiovascular outcomes to sleep staging, across devices and body locations.
- Published: arXiv preprint, 2026
- Signals: ECG, PPG
- Method: masked cross-modal reconstruction
Models
Datasets used in this work
- Samsung AFib/PVC study: Galaxy Watch PPG with Preventice ECG-patch beat labels (private)
- Samsung BP free-living study: Galaxy Watch PPG with cuff-based BP (private)
- Samsung BP lab study: Galaxy Watch PPG with CNAP continuous finger BP reference (private)
- Samsung Galaxy Watch PPG pretraining corpus (~32K users, ~17M 10-s segments; private)