IMC-PINN-FE: A Physics-Informed Neural Network for Patient-Specific Left Ventricular Finite Element Modeling with Image Motion Consistency and Biomechanical Parameter Estimation

2025-06-25

AI Summary

Traditional finite element (FE) methods face two major bottlenecks in patient-specific left ventricular biomechanical modeling: (1) difficulty in directly inferring key material parameters—such as myocardial stiffness and active tension—from clinical imaging, and (2) prohibitively high computational cost and low motion-matching accuracy over the full cardiac cycle. To address these challenges, we propose an end-to-end framework integrating physics-informed neural networks (PINNs), attention-based motion estimation, and an unsupervised cyclic regularization network. This framework embeds image-based motion consistency constraints and inverse material parameter estimation directly into the FE simulation pipeline, requiring only single-subject cine MRI and intraventricular pressure data for joint parameter optimization. Our method accelerates inverse FE computation to the second level and achieves 75× speedup for full-cycle simulation. Dice score improves from 0.849 to 0.927, and pressure–volume relationship reconstruction accuracy is significantly enhanced. To our knowledge, this is the first approach enabling high-fidelity, computationally efficient, and fully data-driven myocardial biomechanical parameter personalization.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Machine Learning: Bio-inspired Learning
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Elucidating the biomechanical behavior of the myocardium is crucial for understanding cardiac physiology, but cannot be directly inferred from clinical imaging and typically requires finite element (FE) simulations. However, conventional FE methods are computationally expensive and often fail to reproduce observed cardiac motions. We propose IMC-PINN-FE, a physics-informed neural network (PINN) framework that integrates imaged motion consistency (IMC) with FE modeling for patient-specific left ventricular (LV) biomechanics. Cardiac motion is first estimated from MRI or echocardiography using either a pre-trained attention-based network or an unsupervised cyclic-regularized network, followed by extraction of motion modes. IMC-PINN-FE then rapidly estimates myocardial stiffness and active tension by fitting clinical pressure measurements, accelerating computation from hours to seconds compared to traditional inverse FE. Based on these parameters, it performs FE modeling across the cardiac cycle at 75x speedup. Through motion constraints, it matches imaged displacements more accurately, improving average Dice from 0.849 to 0.927, while preserving realistic pressure-volume behavior. IMC-PINN-FE advances previous PINN-FE models by introducing back-computation of material properties and better motion fidelity. Using motion from a single subject to reconstruct shape modes also avoids the need for large datasets and improves patient specificity. IMC-PINN-FE offers a robust and efficient approach for rapid, personalized, and image-consistent cardiac biomechanical modeling.

Problem

• Estimates myocardial stiffness and active tension rapidly
• Accelerates FE modeling for cardiac biomechanics significantly
• Improves motion fidelity and patient-specific modeling accuracy

Innovation

• Physics-informed neural network for cardiac modeling
• Image motion consistency with finite element modeling
• Rapid estimation of myocardial stiffness and tension