New Publication: Optimizing Noninvasive FFR with Machine Learning and Efficient Hemodynamic Modeling

In this work published in Cardiovascular Engineering and Technology, we present a scalable approach for estimating fractional flow reserve (FFR) by combining steady-state 1D hemodynamic modeling with machine learning refinement. By reducing computational complexity while leveraging clinical data, the method achieves strong agreement with invasive measurements and high diagnostic accuracy, supporting practical deployment of noninvasive, patient-specific cardiovascular models.