Learning Disease-Sensitive Latent Interaction Graphs From Noisy Cardiac Flow Measurements

2026-02-26

AI Summary

Existing methods struggle to extract disease-relevant coherent vortex structures and their dynamic interactions from noisy cardiac blood flow data. This work proposes the first physics-informed latent graph learning framework that models cardiac vortices as interacting nodes, integrating neural relational inference, a physics-inspired interaction energy model, and vortex birth–death dynamics to enable interpretable, cross-modal modeling across computational fluid dynamics (CFD) and ultrasound data. The introduced graph entropy metric exhibits strong correlation with disease severity: in coarctation of the aorta simulations, graph entropy significantly correlates with stenosis severity (R² = 0.78, Spearman |ρ| = 0.96); in left ventricular ultrasound data, the method successfully captures vortex weakening induced by assist devices, demonstrating its generalizability and clinical potential.

Technology Category

Machine Learning: Graph-based Machine LearningReasoning under Uncertainty: Graphical ModelsComputer Vision: Low Level & Physics-based Vision

Application Category

Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphsSemantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMsSearch and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Cardiac blood flow patterns contain rich information about disease severity and clinical interventions, yet current imaging and computational methods fail to capture underlying relational structures of coherent flow features. We propose a physics-informed, latent relational framework to model cardiac vortices as interacting nodes in a graph. Our model combines a neural relational inference architecture with physics-inspired interaction energy and birth-death dynamics, yielding a latent graph sensitive to disease severity and intervention level. We first apply this to computational fluid dynamics simulations of aortic coarctation. Learned latent graphs reveal that as the aortic radius narrows, vortex interactions become stronger and more frequent. This leads to a higher graph entropy, correlating monotonically with coarctation severity ($R^2=0.78$, Spearman $|ρ|=0.96$). We then extend this method to ultrasound datasets of left ventricles under varying levels of left ventricular assist device support. Again the latent graph representation captures the weakening of coherent vortical structures, thereby demonstrating cross-modal generalisation. Results show latent interaction graphs and entropy serve as robust and interpretable markers of cardiac disease and intervention.

Problem

• cardiac flow
• latent interaction graphs
• disease severity
• noisy measurements
• vortex dynamics

Innovation

• physics-informed graph learning
• latent interaction graphs
• cardiac vortex dynamics
• neural relational inference
• graph entropy