Hybrid Model Learning for Cardiovascular Biomarkers Inference

This paper was accepted at the workshop Deep Generative Models for Health at NeurIPS 2023.

Cardiovascular diseases (CVDs) are a major global health concern, making the longitudinal monitoring of cardiovascular biomarkers vital for early diagnosis and intervention. A core challenge is the inference of cardiac pulse parameters from pulse waves, especially when acquired from wearable sensors at peripheral body locations. Traditional machine learning (ML) approaches face hurdles in this context due to the scarcity of labeled data, primarily sourced from clinical settings. Simultaneously, physical models, like the whole-body 1D hemodynamics simulators, although informative, struggle with the inverse problem and the complications posed by parameter interactions. Recent work has turned to simulation-based inference (SBI) to inform parameter inference by leveraging model simulations. Still, transferring predictors from simulations to real-world data remains a challenge due to model misspecifications. Addressing these issues, this paper presents a novel hybrid learning approach. By fusing a pulse-wave propagation simulator with a data-driven correction model, our methodology aims to blend the rigor of physical models with the flexibility of ML, offering a promising avenue for effective cardiovascular biomarker monitoring.