Key result
Patient-specific identification of aortic pulse wave velocity via a neural network predicted brachial-ankle AoPWV with an RMSE of 1.3 m/s and improved FFR estimation error from 4.4% to 3.8%.
Why the study?
Personalized numerical simulation of coronary hemodynamic indices requires accurate identification of vessel elasticity, but direct measurement of coronary elasticity is not clinically available.
Does a neural network approach for estimating patient-specific AoPWV improve the accuracy of simulated fractional flow reserve (FFR) compared to using a constant AoPWV?
Population
Synthetic database of virtual subjects and data from real patients
Comparison
Simulated hemodynamic indices using predicted AoPWV vs constant AoPWV (7.5 m/s)
Design
Computational simulation and neural network validation study
Authors
Loading...
Patient-specific AoPWV estimation may modestly refine simulated FFR; leaves open prospective clinical validation before adoption.
Does a neural network approach for estimating patient-specific AoPWV improve the accuracy of simulated fractional flow reserve (FFR) compared to using a constant AoPWV?
Absolute Event Rate: 3.8% vs 4.4%
A neural network using basic clinical parameters can accurately estimate patient-specific aortic pulse wave velocity, which improves the computational simulation of fractional flow reserve.
Гамилов et al. (2023) studied Coronary stenosis. Neural network approach for estimating aortic pulse wave velocity (AoPWV) vs. Constant AoPWV (7.5 m/s) was evaluated on Estimation error of fractional flow reserve (FFR). Patient-specific identification of aortic pulse wave velocity via a neural network predicted brachial-ankle AoPWV with an RMSE of 1.3 m/s and improved FFR estimation error from 4.4% to 3.8%.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: