Shape-informed cardiac mechanics surrogates in data-scarce regimes via geometric encoding and generative augmentation

2026-02-23

AI Summary

High-fidelity cardiac mechanical simulations are computationally expensive and challenging to deploy clinically, while existing surrogate models exhibit limited generalization under anatomical variability and data scarcity. This work proposes a two-stage framework: first, a compact latent representation of left ventricular geometry is learned via PCA and DeepSDF to generate synthetic data that augments the training set; subsequently, a neural field–based conditional surrogate model is constructed, leveraging universal ventricular coordinates for displacement prediction. By decoupling geometric representation from physical response learning, the approach significantly enhances generalization across diverse anatomies. Experiments demonstrate that the model achieves high predictive accuracy on both idealized and patient-specific datasets, exhibiting strong robustness to unseen anatomies, sparse inputs, and noisy observations.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

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

Abstract

High-fidelity computational models of cardiac mechanics provide mechanistic insight into the heart function but are computationally prohibitive for routine clinical use. Surrogate models can accelerate simulations, but generalization across diverse anatomies is challenging, particularly in data-scarce settings. We propose a two-step framework that decouples geometric representation from learning the physics response, to enable shape-informed surrogate modeling under data-scarce conditions. First, a shape model learns a compact latent representation of left ventricular geometries. The learned latent space effectively encodes anatomies and enables synthetic geometries generation for data augmentation. Second, a neural field-based surrogate model, conditioned on this geometric encoding, is trained to predict ventricular displacement under external loading. The proposed architecture performs positional encoding by using universal ventricular coordinates, which improves generalization across diverse anatomies. Geometric variability is encoded using two alternative strategies, which are systematically compared: a PCA-based approach suitable for working with point cloud representations of geometries, and a DeepSDF-based implicit neural representation learned directly from point clouds. Overall, our results, obtained on idealized and patient-specific datasets, show that the proposed approaches allow for accurate predictions and generalization to unseen geometries, and robustness to noisy or sparsely sampled inputs.

Problem

• cardiac mechanics
• surrogate modeling
• data-scarce regimes
• geometric generalization
• shape representation

Innovation

• geometric encoding
• generative augmentation
• neural field surrogate
• universal ventricular coordinates
• data-scarce regimes