Leveraging Cardiovascular Simulations for In-Vivo Prediction of Cardiac Biomarkers

TL;DR

This paper presents a novel simulation-based inference framework using neural networks to predict cardiac biomarkers from arterial waveforms, validated both in silico and in vivo, improving uncertainty quantification and measurement reliability.

Contribution

It introduces a new neural posterior estimation method trained on a large cardiac simulation dataset, incorporating stochastic effects to better match real-world data and enable accurate biomarker prediction.

Findings

Accurately predicts Heart Rate, Cardiac Output, SVR, and LVET from waveforms.

Validates predictions with in vivo data showing reliable temporal trend capture.

Uncertainty estimates correlate with prediction errors, enabling automatic measurement rejection.

Abstract

Whole-body hemodynamics simulators, which model blood flow and pressure waveforms as functions of physiological parameters, are now essential tools for studying cardiovascular systems. However, solving the corresponding inverse problem of mapping observations (e.g., arterial pressure waveforms at specific locations in the arterial network) back to plausible physiological parameters remains challenging. Leveraging recent advances in simulation-based inference, we cast this problem as statistical inference by training an amortized neural posterior estimator on a newly built large dataset of cardiac simulations that we publicly release. To better align simulated data with real-world measurements, we incorporate stochastic elements modeling exogenous effects. The proposed framework can further integrate in-vivo data sources to refine its predictive capabilities on real-world data. In silico, we demonstrate that the proposed framework enables finely quantifying uncertainty associated with individual measurements, allowing trustworthy prediction of four biomarkers of clinical interest--namely Heart Rate, Cardiac Output, Systemic Vascular Resistance, and Left Ventricular Ejection Time--from arterial pressure waveforms and photoplethysmograms. Furthermore, we validate the framework in vivo, where our method accurately captures temporal trends in CO and SVR monitoring on the VitalDB dataset. Finally, the predictive error made by the model monotonically increases with the predicted uncertainty, thereby directly supporting the automatic rejection of unusable measurements.