用物理模型提升可穿戴设备心率信号的临床解释性
Inferring Optical Tissue Properties from Photoplethysmography using Hybrid Amortized Inference
- 结合生物物理模型与深度学习，实现从脉搏波信号推断生理参数
- 在多种噪声和传感器条件下仍能准确估计生理参数
- 适合需要可解释性医疗算法的研究者与硬件设计师
智能可穿戴设备可通过光电容积脉搏波描记法（PPG）持续监测心率、心率变异性及血氧饱和度等指标。近年来深度学习研究发现，PPG波形还蕴含更丰富的生理信息。然而，现有深度学习模型依赖于缺乏明确生理意义的特征，导致预测性能与临床可解释性、传感器设计之间存在矛盾。本文提出PPGen，一个将PPG信号与可解释的生理与光学参数关联的生物物理模型。在此基础上，我们进一步设计混合摊销推断（HAI）方法，实现从PPG信号中快速、鲁棒且可扩展地估计关键生理参数，并能纠正模型误设问题。大规模仿真实验表明，HAI在不同噪声水平和传感器配置下均能准确推断生理参数。结果展示了兼具深度学习特征保真度与临床可解释性的未来方向，支持更明智的硬件设计。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。