A physiologically constrained calibration framework for cardiovascular models applied to a paediatric synthetic sepsis population
4 affiliations
- 1Department of Mechanical Engineering, University College London, London, United Kingdom.
- 2Hawkes Institute, University College London, London, United Kingdom.
- 3Paediatric Intensive Care Unit, Great Ormond Street Hospital for Children NHS Foundation Trust, London, United Kingdom.
- 4Department of Mathematics, University College London, London, United Kingdom.
Abstract
Calibration of mechanistic cardiovascular models is a central barrier to their use in population analysis and patient-specific simulation, particularly in settings where key physiological variables are unobservable and multiple parameter combinations can reproduce the same haemodynamic targets. In this work, we present Embedded Feedback Controller (EFC), a calibration framework for ODE-based lumped-parameter cardiovascular models in which selected physiological parameters are promoted to dynamic states and driven toward prescribed targets through embedded controller equations. By exploiting the qualitative structure of the governing equations, EFC enforces physiologically consistent parameter-variable relationships and converges to the single solution admitted by an over-determined set of targets, reproducibly and independently of the initial conditions, at a cost that scales efficiently with model complexity. The framework is demonstrated in silico, using a mechanistic cardiovascular model to generate virtual paediatric populations spanning normal physiology and two septic shock phenotypes (warm and cold shock) from literature-derived target ranges, achieving <1% residual error across pressures, flows, and compartmental volumes. The resulting parameter distributions are consistent, by construction with theoretical haemodynamic adaptations in paediatric sepsis, including alterations in vascular resistance, compliance, cardiac elastance, and effective blood volume. Importantly, persistent calibration residuals arise when target combinations are structurally incompatible with the model and the parameter carrying the residual is held at a bound, providing an explicit and interpretable diagnostic of feasibility limits. These results establish EFC as a general, scalable calibration strategy for mechanistic cardiovascular models and a practical foundation for virtual population generation and future patient-specific digital twin applications in critical care. Author summary: We present a new way to calibrate mechanistic cardiovascular models by updating selected model parameters during simulation while preserving physiological relationships. This makes calibration robust to initial conditions and supports scalable generation of mechanistically interpretable virtual populations. We demonstrate the method in a closed-loop paediatric cardiovascular model and reproduce normal circulation and warm and cold septic shock phenotypes with low residual error and plausible parameter shifts. Residual mismatches remain diagnostically meaningful, flagging infeasible target combinations.
Related
- Physiologically Based Pharmacokinetic Modeling to Predict Drug-Drug Interactions of Aficamten in the Chinese Population.
- Presence of Dolosigranulum pigrum in the nasopharynx and its relationship with respiratory health status in paediatric population
- Population pharmacokinetics of cefazolin administered as prophylaxis in paediatric patients undergoing cardiac surgery
- A unified population pharmacokinetic model of amikacin in paediatric patients from birth to 18 years for model-informed precision dosing.
- Role of Hyperbilirubinaemia as a Predictor of Complicated Appendicitis in Paediatric Population
BibTeX
@article{cabeleira2026physiologically,
title = {{A physiologically constrained calibration framework for cardiovascular models applied to a paediatric synthetic sepsis population}},
author = {Manuel T Cabeleira and Samiran Ray and Nicholas C Ovenden and Vanessa Díaz-Zuccarini},
journal = {PLOS Computational Biology},
year = {2026},
month = oct,
doi = {10.1371/journal.pcbi.1014847},
url = {https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014847},
note = {CC BY},
}
RIS
TY - JOUR
TI - A physiologically constrained calibration framework for cardiovascular models applied to a paediatric synthetic sepsis population
AU - Manuel T Cabeleira
AU - Samiran Ray
AU - Nicholas C Ovenden
AU - Vanessa Díaz-Zuccarini
PY - 2026
DA - 2026/10/07
JO - PLOS Computational Biology
DO - 10.1371/journal.pcbi.1014847
UR - https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014847
AB - Calibration of mechanistic cardiovascular models is a central barrier to their use in population analysis and patient-specific simulation, particularly in settings where key physiological variables are unobservable and multiple parameter combinations can reproduce the same haemodynamic targets. In this work, we present Embedded Feedback Controller (EFC), a calibration framework for ODE-based lumped-parameter cardiovascular models in which selected physiological parameters are promoted to dynamic states and driven toward prescribed targets through embedded controller equations. By exploiting the qualitative structure of the governing equations, EFC enforces physiologically consistent parameter-variable relationships and converges to the single solution admitted by an over-determined set of targets, reproducibly and independently of the initial conditions, at a cost that scales efficiently with model complexity. The framework is demonstrated in silico, using a mechanistic cardiovascular model to generate virtual paediatric populations spanning normal physiology and two septic shock phenotypes (warm and cold shock) from literature-derived target ranges, achieving <1% residual error across pressures, flows, and compartmental volumes. The resulting parameter distributions are consistent, by construction with theoretical haemodynamic adaptations in paediatric sepsis, including alterations in vascular resistance, compliance, cardiac elastance, and effective blood volume. Importantly, persistent calibration residuals arise when target combinations are structurally incompatible with the model and the parameter carrying the residual is held at a bound, providing an explicit and interpretable diagnostic of feasibility limits. These results establish EFC as a general, scalable calibration strategy for mechanistic cardiovascular models and a practical foundation for virtual population generation and future patient-specific digital twin applications in critical care. Author summary: We present a new way to calibrate mechanistic cardiovascular models by updating selected model parameters during simulation while preserving physiological relationships. This makes calibration robust to initial conditions and supports scalable generation of mechanistically interpretable virtual populations. We demonstrate the method in a closed-loop paediatric cardiovascular model and reproduce normal circulation and warm and cold septic shock phenotypes with low residual error and plausible parameter shifts. Residual mismatches remain diagnostically meaningful, flagging infeasible target combinations.
SN - 1553-7358
SN - 1553-734X
ER -