Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Viraj Patel, Marko Grujic, Philipp Aigner, Theodor Abart, Marcus Granegger, Deblina Bhattacharjee, Katharine Fraser

TL;DR

This 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 and results show latent interaction graphs and entropy serve as robust and interpretable markers of cardiac disease and intervention.

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.

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

TL;DR

This 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 and results show latent interaction graphs and entropy serve as robust and interpretable markers of cardiac disease and intervention.

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 (, Spearman ). 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.
Paper Structure (13 sections, 5 equations, 2 figures, 1 table)

This paper contains 13 sections, 5 equations, 2 figures, 1 table.

Table of Contents

  1. Background and Motivation
  2. Introduction
  3. Neural Relational Inference
  4. Method
  5. Data Preparation
  6. Latent Interaction Model
  7. Implementation
  8. Results and Discussion
  9. CFD: Aortic Coarctation
  10. Ultrasound: LVAD Supported Ventricle
  11. Ablations
  12. Limitations and Future Work
  13. Conclusion

Figures (2)

  • Figure 1: NRI architecture, with modifications compared with kipfNeuralRelationalInference2018 in orange. The encoder, which used a disease severity conditioned physics head, and decoder, are alternating MLPs and message passing layers. Solid blocks represent node-level embeddings and multi-coloured blocks represent edge-level embeddings.
  • Figure 2: Aortic coarctation: (a) the distribution of edge probability and (b) the entropy of the latent interaction graph changes with coarctation level. LVAD support: (c) the distribution of edge probability (for no interaction) and (d) the entropy of the latent interaction graph changes with LVAD support level. Decreasing % of aorta radius corresponds to increasing disease severity. LVAD support levels: 0.0 = no support, 0.5 = partial support, and 1.0 = full support.