AuthorsJens Behrmann‡, Maria Cervera‡, Antoine Wehenkel‡, Andrew C. Miller, Albert Cerussi, Pranay Jain, Vivek Venugopal, Shijie Yan, Guillermo Sapiro, Luca Pegolotti†**, Jörn-Henrik Jacobsen†**
Smart wearables enable continuous tracking of established biomarkers such as heart rate, heart rate variability, and blood oxygen saturation via photoplethysmography (PPG). Beyond these metrics, PPG waveforms contain richer physiological information, as recent deep learning (DL) studies demonstrate. However, DL models often rely on features with unclear physiological meaning, creating a tension between predictive power, clinical interpretability, and sensor design. We address this gap by introducing PPGen, a biophysical model that relates PPG signals to interpretable physiological and …
AuthorsJens Behrmann‡, Maria Cervera‡, Antoine Wehenkel‡, Andrew C. Miller, Albert Cerussi, Pranay Jain, Vivek Venugopal, Shijie Yan, Guillermo Sapiro, Luca Pegolotti†**, Jörn-Henrik Jacobsen†**
Smart wearables enable continuous tracking of established biomarkers such as heart rate, heart rate variability, and blood oxygen saturation via photoplethysmography (PPG). Beyond these metrics, PPG waveforms contain richer physiological information, as recent deep learning (DL) studies demonstrate. However, DL models often rely on features with unclear physiological meaning, creating a tension between predictive power, clinical interpretability, and sensor design. We address this gap by introducing PPGen, a biophysical model that relates PPG signals to interpretable physiological and optical parameters. Building on PPGen, we propose hybrid amortized inference (HAI), enabling fast, robust, and scalable estimation of relevant physiological parameters from PPG signals while correcting for model misspecification. In extensive in-silico experiments, we show that HAI can accurately infer physiological parameters under diverse noise and sensor conditions. Our results illustrate a path toward PPG models that retain the fidelity needed for DL-based features while supporting clinical interpretation and informed hardware design.
- † Isomorphic Labs
- ** Work done while at Apple
- ‡ Shared first authorship
Related readings and updates.
This paper was accepted at the workshop Machine Learning and the Physical Sciences at NeurIPS 2023.
Over the past decades, hemodynamics simulators have steadily evolved and have become tools of choice for studying cardiovascular systems in-silico. This comes naturally at the cost of increasing complexity since state-of-the-art models are non-linear partial differential equations depending on many parameters. While such tools are routinely used…
The abundance of wrist-worn heart rate measuring devices enables long term cardiovascular monitoring through photoplethysmography (PPG). Such signals contain unique identifiable information that can help in biometric authentication. In this work, we propose Fusion-ID, which use wrist-worn PPG sensors fused with motion sensor data as a way to do bio authentication on wrist worn devices. We conducted a user study using a PPG and motion sensor…