From Hemodynamics to Energetics: Parameter Identification in a Closed-Loop Model of the Cardiovascular System.

Heart failure is a complex condition, and accurately stratifying patients using standard hemodynamic assessments remains a challenging task. To address this, sophisticated scoring systems have been developed to better capture the interplay between loading conditions, energetics, and cardiac performance. However, applying these scores in clinical practice is difficult, as they are often calculated indirectly from measurable variables and depend on numerous assumptions. In this work, we propose a model calibration procedure leveraged by the covariance matrix adaptation evolution strategy (CMA-ES) to estimate model parameters in a closed-loop compartmental model of the cardiovascular system based on hemodynamic patient-specific data. The goal of this procedure is to reverse-engineer the patient-specific cardiovascular model, enabling us to reveal the patient's cardiac energetics, and estimate the scores that better characterize the patient's physiologic state. We propose different scenarios based on data availability and investigate model identification under these conditions. Understanding the patient's energetic profile is crucial for gaining insight into the underlying physiologic and pathophysiologic conditions, as well as for tailoring patient-specific therapeutic interventions effectively. We demonstrate the capabilities of this methodology using a cohort of 20 heart failure patients. The effectiveness of the proposed approach in calibrating patient-specific models to match hemodynamic measurements obtained in the catheterization laboratory is analyzed for the different data availability scenarios. This enables a hemodynamically reliable, individualized characterization of the energetic profile, serving as a valuable clinical decision support tool for guiding heart failure treatment.