Physics-Guided Variational Method for Fractional Flow Reserve Based on Coronary Angiography

Area of Science:

• Cardiovascular Medicine
• Biomedical Engineering
• Computational Fluid Dynamics

Background:

• Coronary ischemia is a major cause of death, necessitating precise diagnostic tools.
• Fractional flow reserve (FFR) combined with coronary angiography aids in assessing coronary stenosis and guiding revascularization.
• Current FFR methods are invasive, and existing non-invasive techniques struggle with complex fluid-structure interactions (FSI).

Purpose of the Study:

• To develop a non-invasive method for estimating FFR using a physics-guided variational domain progressing method (PVDPM).
• To accurately model the fluid-structure interaction (FSI) in coronary arteries for improved FFR estimation.
• To provide a reliable diagnostic solution for coronary ischemia based on coronary angiography.

Main Methods:

• Proposed a physics-guided variational domain progressing method (PVDPM) to model the FSI system.
• Utilized the principle of virtual work for comprehensive FSI modeling.
• Integrated coronary angiography-derived vascular morphology with biomechanical principles.

Main Results:

• The PVDPM achieved 91% accuracy on clinical datasets.
• Successfully modeled the complex FSI of coronary flow and vessel walls.
• Demonstrated the capability for accurate, non-invasive FFR estimation.

Conclusions:

• The PVDPM offers an accurate and non-invasive approach for FFR estimation.
• This method enhances the diagnostic capabilities for coronary ischemia using coronary angiography.
• PVDPM addresses the limitations of traditional invasive FFR measurements and complex modeling challenges.