Unrolled Reconstruction with Integrated Super-Resolution for Accelerated 3D LGE MRI

2026-03-18

AI Summary

This work addresses the challenge of high-frequency detail loss in thin atrial structures during accelerated 3D late gadolinium enhancement (LGE) MRI reconstruction from undersampled k-space data. To this end, the authors propose a hybrid unrolled reconstruction framework that, for the first time, directly integrates a deep super-resolution prior—implemented via an EDSR network—into a model-based iterative optimization pipeline, replacing conventional proximal operators. At each iteration, the method jointly enforces data consistency and super-resolution enhancement, enabling end-to-end joint optimization of image reconstruction and resolution refinement. Experimental results demonstrate that the proposed approach consistently outperforms existing methods—including compressed sensing, MoDL, and self-guided DIP—across multiple acceleration factors, achieving significant improvements in PSNR, SSIM, and left atrial segmentation accuracy.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Mixed Discrete/Continuous Search
• Computer Vision: 3D Computer Vision

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• 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: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Accelerated 3D late gadolinium enhancement (LGE) MRI requires robust reconstruction methods to recover thin atrial structures from undersampled k-space data. While unrolled model-based networks effectively integrate physics-driven data consistency with learned priors, they operate at the acquired resolution and may fail to fully recover high-frequency detail. We propose a hybrid unrolled reconstruction framework in which an Enhanced Deep Super-Resolution (EDSR) network replaces the proximal operator within each iteration of the optimization loop, enabling joint super-resolution enhancement and data consistency enforcement. The model is trained end-to-end on retrospectively undersampled preclinical 3D LGE datasets and compared against compressed sensing, Model-Based Deep Learning (MoDL), and self-guided Deep Image Prior (DIP) baselines. Across acceleration factors, the proposed method consistently improves PSNR and SSIM over standard unrolled reconstruction and better preserves fine cardiac structures, leading to improved LA (left atrium) segmentation performance. These results demonstrate that integrating super-resolution priors directly within model-based reconstruction provides measurable gains in accelerated 3D LGE MRI.

Problem

• 3D LGE MRI
• accelerated MRI
• super-resolution
• undersampled k-space
• atrial structures

Innovation

• unrolled reconstruction
• super-resolution
• model-based deep learning
• 3D LGE MRI
• EDSR