A multi-dynamic low-rank deep image prior (ML-DIP) for 3D real-time cardiovascular MRI

TL;DR

This paper introduces ML-DIP, a novel deep learning framework that reconstructs high-quality 3D real-time cardiac MRI from highly undersampled data without needing fully sampled training data, enabling faster imaging with preserved motion details.

Contribution

ML-DIP models spatial and deformation information with separate neural networks, jointly trained per scan, to improve 3D cardiac MRI reconstruction from undersampled data without external training datasets.

Findings

Achieved PSNR > 29 dB and SSIM > 0.90 in phantom studies with 2-minute scans.

Provided functional cardiac measurements comparable to 2D cine in vivo.

Reconstructed irregular beats and motion artifacts better than existing methods.

Abstract

Purpose: To develop a reconstruction framework for 3D real-time cine cardiovascular magnetic resonance (CMR) from highly undersampled data without requiring fully sampled training datasets. Methods: We developed a multi-dynamic low-rank deep image prior (ML-DIP) framework that models spatial image content and deformation fields using separate neural networks. These sub-networks are jointly trained per scan to reconstruct the dynamic image series directly from undersampled k-space data. ML-DIP was evaluated on (i) a 3D cine digital phantom with simulated premature ventricular contractions (PVCs), (ii) ten healthy subjects (including two scanned during both rest and exercise), and (iii) 12 patients with a history of PVCs. Phantom results were assessed using peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM). In vivo performance was evaluated by comparing left-ventricular function quantification (against 2D real-time cine) and image quality (against 2D real-time cine and binning-based 5D-Cine). Results: In the phantom study, ML-DIP achieved PSNR > 29 dB and SSIM > 0.90 for scan times as short as two minutes, while recovering cardiac motion, respiratory motion, and PVC events. In healthy subjects, ML-DIP yielded functional measurements comparable to 2D cine and higher image quality than 5D-Cine, including during exercise with high heart rates and bulk motion. In PVC patients, ML-DIP preserved beat-to-beat variability and reconstructed irregular beats, whereas 5D-Cine showed motion artifacts and information loss due to binning. Conclusion: ML-DIP enables high-quality 3D real-time CMR with acceleration factors exceeding 1,000 by learning low-rank spatial and motion representations from undersampled data, without relying on external fully sampled training datasets.