Learning Geometry-Dependent and Physics-Based Inverse Image Reconstruction

Abstract

Deep neural networks have shown great potential in image reconstruction problems in Euclidean space. However, many reconstruction problems involve imaging physics that are dependent on the underlying non-Euclidean geometry. In this paper, we present a new approach to learn inverse imaging that exploit the underlying geometry and physics. We first introduce a non-Euclidean encoding-decoding network that allows us to describe the unknown and measurement variables over their respective geometrical domains. We then learn the geometry-dependent physics in between the two domains by explicitly modeling it via a bipartite graph over the graphical embedding of the two geometry. We applied the presented network to reconstructing electrical activity on the heart surface from body-surface potential. In a series of generalization tasks with increasing difficulty, we demonstrated the improved ability of the presented network to generalize across geometrical changes underlying the data in comparison to its Euclidean alternatives.

References in corpus (6)

• A Comprehensive Survey on Graph Neural Networks
• Geometric deep learning: going beyond Euclidean data
• Image reconstruction by domain transform manifold learning
• Learned Primal-dual Reconstruction
• Generative Modeling and Inverse Imaging of Cardiac Transmembrane Potential
• Bayesian Optimization on Large Graphs via a Graph Convolutional Generative Model: Application in Cardiac Model Personalization

Cited by in corpus (1)

• Few-shot Generation of Personalized Neural Surrogates for Cardiac Simulation via Bayesian Meta-Learning