Key result
Mechanistic modeling integration improves Fontan outcome prediction accuracy to an AUROC of ~0.78.
Why the study?
Does the integration of mechanistic Fontan circulatory models with machine learning improve the prediction of patient outcomes compared to clinical-only approaches in patients with Fontan circulation?
Population
51 patients with Fontan circulation, age 25±8 years
Comparison
Integration of physics-based hemodynamic… vs Clinical-only approaches
Design
Other
Authors
Loading...
Integrating physics-based hemodynamic modeling with machine learning improves risk stratification and prediction of adverse outcomes in patients with Fontan circulation compared to clinical metrics alone.
Observational (n=51)
Does the integration of mechanistic Fontan circulatory models with machine learning improve the prediction of patient outcomes compared to clinical-only approaches in patients with Fontan circulation?
Integrating physics-based hemodynamic modeling with machine learning improves risk stratification and prediction of adverse outcomes in patients with Fontan circulation compared to clinical metrics alone.
Schenk et al. (2025) conducted an observational in Fontan circulation (n=51). Model-informed classifiers (lumped-parameter mechanistic model) vs. Clinical-only classifiers was evaluated on Patient outcomes (listing for heart transplant and/or cardiac death). Integration of lumped-parameter mechanistic model parameters increased the AUROC for predicting Fontan patient outcomes (transplant listing or cardiac death) from 0.67 to 0.78.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: