Physics-informed self-supervised learning for predictive modeling of coronary artery digital twins

2025-11-25

AI Summary

Cardiovascular disease—particularly coronary artery disease (CAD)—demands scalable, interpretable methods for early risk prediction. Existing 3D digital twin–based computational fluid dynamics (CFD) modeling is computationally prohibitive, while purely data-driven approaches suffer from scarce labeled data and lack of physiological priors. To address this, we propose a physics-guided self-supervised graph neural network (GNN) framework: it incorporates the 1D Navier–Stokes equations and pressure-drop law into GNN pretraining, enabling physiology-consistent representation learning without CFD simulations or ground-truth labels. Coupled with synthetic digital twin generation and fine-tuning on multicenter clinical data, the framework accurately models spatially resolved pressure and fractional flow reserve (FFR). Evaluated on 635 patients from the FAME2 trial, our method achieves an AUC of 0.73 for cardiovascular event prediction—significantly outperforming conventional clinical scores—and delivers interpretable, quantitative biomarkers.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Graphical Models
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Cardiovascular disease is the leading global cause of mortality, with coronary artery disease (CAD) as its most prevalent form, necessitating early risk prediction. While 3D coronary artery digital twins reconstructed from imaging offer detailed anatomy for personalized assessment, their analysis relies on computationally intensive computational fluid dynamics (CFD), limiting scalability. Data-driven approaches are hindered by scarce labeled data and lack of physiological priors. To address this, we present PINS-CAD, a physics-informed self-supervised learning framework. It pre-trains graph neural networks on 200,000 synthetic coronary digital twins to predict pressure and flow, guided by 1D Navier-Stokes equations and pressure-drop laws, eliminating the need for CFD or labeled data. When fine-tuned on clinical data from 635 patients in the multicenter FAME2 study, PINS-CAD predicts future cardiovascular events with an AUC of 0.73, outperforming clinical risk scores and data-driven baselines. This demonstrates that physics-informed pretraining boosts sample efficiency and yields physiologically meaningful representations. Furthermore, PINS-CAD generates spatially resolved pressure and fractional flow reserve curves, providing interpretable biomarkers. By embedding physical priors into geometric deep learning, PINS-CAD transforms routine angiography into a simulation-free, physiology-aware framework for scalable, preventive cardiology.

Problem

• Predicts cardiovascular events using physics-informed deep learning
• Eliminates need for CFD in coronary artery digital twin analysis
• Generates interpretable biomarkers from routine angiography without labeled data

Innovation

• Physics-informed self-supervised learning for coronary artery modeling
• Pre-trains graph neural networks on synthetic data using fluid equations
• Generates interpretable biomarkers without computational fluid dynamics simulations