![[2602.16000] Imaging-Derived Coronary Fractional Flow Reserve: Advances in Physics-Based, Machine-Learning, and Physics-Informed Methods](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.16000] Imaging-Derived Coronary Fractional Flow Reserve: Advances in Physics-Based, Machine-Learning, and Physics-Informed Methods
Summary
This article reviews advances in imaging-derived fractional flow reserve (FFR) methods, focusing on machine learning and physics-informed approaches for assessing coronary stenosis.
Why It Matters
The integration of machine learning and physics-informed methods in assessing coronary stenosis represents a significant advancement in medical imaging. These innovations promise faster, more reliable diagnostics, which could enhance patient outcomes and streamline clinical workflows. Understanding these developments is crucial for healthcare professionals and researchers aiming to improve cardiovascular care.
Key Takeaways
- Imaging-derived FFR is evolving with machine learning and physics-informed methods.
- Recent advances improve the speed and automation of coronary assessments.
- Physics-informed neural networks enhance model generalizability and reduce supervision needs.
- Real-world performance of ML approaches can vary due to data heterogeneity.
- Standardized evaluation metrics are essential for safe clinical adoption.
Physics > Medical Physics arXiv:2602.16000 (physics) [Submitted on 17 Feb 2026] Title:Imaging-Derived Coronary Fractional Flow Reserve: Advances in Physics-Based, Machine-Learning, and Physics-Informed Methods Authors:Tanxin Zhu, Emran Hossen, Chen Zhao, Michele Esposito, Jiguang Sun, Weihua Zhou View a PDF of the paper titled Imaging-Derived Coronary Fractional Flow Reserve: Advances in Physics-Based, Machine-Learning, and Physics-Informed Methods, by Tanxin Zhu and 5 other authors View PDF Abstract:Purpose of Review Imaging derived fractional flow reserve (FFR) is rapidly evolving beyond conventional computational fluid dynamics (CFD) based pipelines toward machine learning (ML), deep learning (DL), and physics informed approaches that enable fast, wire free, and scalable functional assessment of coronary stenosis. This review synthesizes recent advances in CT and angiography based FFR, with particular emphasis on emerging physics informed neural networks and neural operators (PINNs and PINOs) and key considerations for their clinical translation. Recent Findings ML/DL approaches have markedly improved automation and computational speed, enabling prediction of pressure and FFR from anatomical descriptors or angiographic contrast dynamics. However, their real-world performance and generalizability can remain variable and sensitive to domain shift, due to multi-center heterogeneity, interpretability challenges, and differences in acquisition protocols and image quality. Ph...
