Context: Computational modeling of cardiovascular disease is increasingly shifting towards comprehensive, multiscale representations of the larger vascular system. Vascular smooth muscle cells (SMCs) play a central role in regulating vascular tone and hemodynamics; responding to several biochemical and physical cues, yet their variability and contribution to system-level behavior is often neglected in current models. Although hypertrophic cardiomyopathy (HCM) is not associated with direct genetic alterations in SMCs, disease-induced changes in hemodynamics and vessel wall properties will often modulate SMC function. Thus, capturing this variability in-silico is important for accurately representing coronary perfusion and cardiovascular dynamics.
Aim: With this work we aim to address biological variability with the development, and extensive validation of computational SMC models that can be integrated into multiscale cardiovascular simulations.
Methods: A population of 10,000 electrophysiological SMC models was generated using latin hypercube sampling to vary key ionic and contractile parameters within physiologically constrained ranges. The models were then calibrated against experimental data and subjected to four distinct physiological challenges; testing numerous SMC electro-chemo-mechanical responses to various stimuli. The test challenged their stability, responsiveness, and functional plausibility. All steps were performed using open source POMtool software.
Results: With all four challenges applied, the low acceptance rate of 2.2% underscores the necessity of rigorous model calibration and validation to ensure physiological plausibility.
Conclusion: The proposed population-of-models framework captures inter-subject variability in SMC electrophysiology and contractility, whilst providing a computationally tractable approach for multiscale integration. Incorporation of these models into coupled vessel and cardiovascular system simulations enables investigation of the impact of SMC variability on coronary perfusion, hemodynamics, and disease phenotypes such as HCM. The framework further supports in-silico evaluation of therapeutic interventions, particularly those targeting calcium handling and contractile dynamics. This work highlights the importance of rigorous model validation and variability quantification in advancing predictive cardiovascular modeling.