Inferring Optical Tissue Properties from Photoplethysmography using Hybrid Amortized Inference
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911189135523840 |
|---|---|
| author | Behrmann, Jens Cervera, Maria R. Wehenkel, Antoine Miller, Andrew C. Cerussi, Albert Jain, Pranay Venugopal, Vivek Yan, Shijie Sapiro, Guillermo Pegolotti, Luca Jacobsen, Jörn-Henrik |
| author_facet | Behrmann, Jens Cervera, Maria R. Wehenkel, Antoine Miller, Andrew C. Cerussi, Albert Jain, Pranay Venugopal, Vivek Yan, Shijie Sapiro, Guillermo Pegolotti, Luca Jacobsen, Jörn-Henrik |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_02073 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Inferring Optical Tissue Properties from Photoplethysmography using Hybrid Amortized Inference Behrmann, Jens Cervera, Maria R. Wehenkel, Antoine Miller, Andrew C. Cerussi, Albert Jain, Pranay Venugopal, Vivek Yan, Shijie Sapiro, Guillermo Pegolotti, Luca Jacobsen, Jörn-Henrik Machine Learning Biological Physics 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. |
| title | Inferring Optical Tissue Properties from Photoplethysmography using Hybrid Amortized Inference |
| topic | Machine Learning Biological Physics |
| url | https://arxiv.org/abs/2510.02073 |