Accelerated MR Elastography Using Learned Neural Network Representation

2026-01-17

AI Summary

This work addresses the challenge of achieving rapid, high-resolution reconstruction in magnetic resonance elastography (MRE) under highly undersampled conditions—such as single-shot spiral trajectories with an acceleration factor of R=10—where high-quality paired training data are typically unavailable. The authors propose a self-supervised deep neural network representation that generalizes the conventional linear subspace model into a nonlinear framework. By integrating multiple physics-informed priors—including multi-level k-space consistency, anatomical similarity, and phase smoothness of harmonic displacements—the method enables high-fidelity reconstruction without requiring fully sampled ground-truth data. Experimental results demonstrate that the proposed approach effectively suppresses noise and artifacts, yielding superior image quality compared to traditional linear subspace methods, while producing elasticity estimates comparable to those obtained from fully sampled data.

Technology Category

Machine Learning: Learning with ManifoldsComputer Vision: Multi-modal VisionIntelligent Robots: State Estimation

Application Category

Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingUser Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

To develop a deep-learning method for achieving fast high-resolution MR elastography from highly undersampled data without the need of high-quality training dataset. We first framed the deep neural network representation as a nonlinear extension of the linear subspace model, then used it to represent and reconstruct MRE image repetitions from undersampled k-space data. The network weights were learned using a multi-level k-space consistent loss in a self-supervised manner. To further enhance reconstruction quality, phase-contrast specific magnitude and phase priors were incorporated, including the similarity of anatomical structures and smoothness of wave-induced harmonic displacement. Experiments were conducted using both 3D gradient-echo spiral and multi-slice spin-echo spiral MRE datasets. Compared to the conventional linear subspace-based approaches, the nonlinear network representation method was able to produce superior image reconstruction with suppressed noise and artifacts from a single in-plane spiral arm per MRE repetition (e.g., total R=10), yielding comparable stiffness estimation to the fully sampled data. This work demonstrated the feasibility of using deep network representations to model and reconstruct MRE images from highly-undersampled data, a nonlinear extension of the subspace-based approaches.

Problem

• MR elastography
• undersampled data
• high-resolution imaging
• fast imaging
• deep learning

Innovation

• deep neural network representation
• self-supervised learning
• MR elastography
• highly undersampled reconstruction
• nonlinear subspace model