Simulation-based Inference for Cardiovascular Models

2023-07-26

arXiv.org

AI Summary

This study addresses the cardiovascular hemodynamics inverse problem—inferring patient-specific physiological parameters from waveform data (e.g., blood pressure, flow)—by introducing the first simulation-based inference (SBI) framework for uncertainty quantification. Methodologically, it integrates a high-fidelity hemodynamic simulator with neural density estimators (e.g., SNPE) to model multidimensional posterior parameter distributions—rather than point estimates—using real clinical waveforms from MIMIC-III. Key contributions are threefold: (1) the first patient-specific posterior uncertainty quantification for five critical biomarkers, including cardiac output and arterial elastance; (2) discovery of subgroup-specific hierarchical structure in measurement uncertainty, surpassing classical sensitivity analysis; and (3) demonstration of high-accuracy heart rate inference and identification of novel noninvasive biomarker estimation capabilities enabled jointly by blood pressure and flow waveforms.

Abstract

Over the past decades, hemodynamics simulators have steadily evolved and have become tools of choice for studying cardiovascular systems in-silico. While such tools are routinely used to simulate whole-body hemodynamics from physiological parameters, solving the corresponding inverse problem of mapping waveforms back to plausible physiological parameters remains both promising and challenging. Motivated by advances in simulation-based inference (SBI), we cast this inverse problem as statistical inference. In contrast to alternative approaches, SBI provides extit{posterior distributions} for the parameters of interest, providing a extit{multi-dimensional} representation of uncertainty for extit{individual} measurements. We showcase this ability by performing an in-silico uncertainty analysis of five biomarkers of clinical interest comparing several measurement modalities. Beyond the corroboration of known facts, such as the feasibility of estimating heart rate, our study highlights the potential of estimating new biomarkers from standard-of-care measurements. SBI reveals practically relevant findings that cannot be captured by standard sensitivity analyses, such as the existence of sub-populations for which parameter estimation exhibits distinct uncertainty regimes. Finally, we study the gap between in-vivo and in-silico with the MIMIC-III waveform database and critically discuss how cardiovascular simulations can inform real-world data analysis.

Problem

• Blood Flow Waveform Analysis
• Physiological Parameter Estimation
• Measurement Uncertainty in Cardiovascular Models

Innovation

• Statistical Body Inference (SBI)
• Quantification of Uncertainty
• New Biomarkers Discovery