English

Physics-Informed Neural ODEs for Temporal Dynamics Modeling in Cardiac T1 Mapping

Image and Video Processing 2025-07-02 v1 Artificial Intelligence

Abstract

Spin-lattice relaxation time (T1T_1) is an important biomarker in cardiac parametric mapping for characterizing myocardial tissue and diagnosing cardiomyopathies. Conventional Modified Look-Locker Inversion Recovery (MOLLI) acquires 11 breath-hold baseline images with interleaved rest periods to ensure mapping accuracy. However, prolonged scanning can be challenging for patients with poor breathholds, often leading to motion artifacts that degrade image quality. In addition, T1T_1 mapping requires voxel-wise nonlinear fitting to a signal recovery model involving an iterative estimation process. Recent studies have proposed deep-learning approaches for rapid T1T_1 mapping using shortened sequences to reduce acquisition time for patient comfort. Nevertheless, existing methods overlook important physics constraints, limiting interpretability and generalization. In this work, we present an accelerated, end-to-end T1T_1 mapping framework leveraging Physics-Informed Neural Ordinary Differential Equations (ODEs) to model temporal dynamics and address these challenges. Our method achieves high-accuracy T1T_1 estimation from a sparse subset of baseline images and ensures efficient null index estimation at test time. Specifically, we develop a continuous-time LSTM-ODE model to enable selective Look-Locker (LL) data acquisition with arbitrary time lags. Experimental results show superior performance in T1T_1 estimation for both native and post-contrast sequences and demonstrate the strong benefit of our physics-based formulation over direct data-driven T1T_1 priors.

Keywords

Cite

@article{arxiv.2507.00613,
  title  = {Physics-Informed Neural ODEs for Temporal Dynamics Modeling in Cardiac T1 Mapping},
  author = {Nuno Capitão and Yi Zhang and Yidong Zhao and Qian Tao},
  journal= {arXiv preprint arXiv:2507.00613},
  year   = {2025}
}

Comments

Submitted version. Accepted at MICCAI 2025

R2 v1 2026-07-01T03:41:18.692Z