Cine cardiac MRI reconstruction using a convolutional recurrent network with refinement
Not on the map. Why?
Abstract (arXiv)
Cine Magnetic Resonance Imaging (MRI) allows for understanding of the heart's function and condition in a non-invasive manner. Undersampling of the k-space is employed to reduce the scan duration, thus increasing patient comfort and reducing the risk of motion artefacts, at the cost of reduced image quality. In this challenge paper, we investigate the use of a convolutional recurrent neural network (CRNN) architecture to exploit temporal correlations in supervised cine cardiac MRI reconstruction. This is combined with a single-image super-resolution refinement module to improve single coil reconstruction by 4.4% in structural similarity and 3.9% in normalised mean square error compared to a plain CRNN implementation. We deploy a high-pass filter to our ℓ1 loss to allow greater emphasis on high-frequency details which are missing in the original data. The proposed model demonstrates considerable enhancements compared to the baseline case and holds promising potential for further improving cardiac MRI reconstruction.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 15 of its 27 references that are in the map. First posted on arXiv 23 Sep 2023.
Builds on
15 of 27 references.
References · 15
- 2017MoDL: Model Based Deep Learning Architecture for Inverse Problemscited by 384
- 2017Learning a Variational Network for Reconstruction of Accelerated MRI Datacited by 340
- 2021Robust Compressed Sensing MRI with Deep Generative Priorscited by 284
- 2017A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstructioncited by 231
- 2019On instabilities of deep learning in image reconstruction - Does AI come at a cost?cited by 211
- 2017Convolutional Recurrent Neural Networks for Dynamic MR Image Reconstructioncited by 105
- 2017Deep learning for undersampled MRI reconstructioncited by 91
- 2018Deep Residual Learning for Accelerated MRI using Magnitude and Phase Networkscited by 53
- 2017MR image reconstruction using deep density priorscited by 37
- 2017Deep Learning with Domain Adaptation for Accelerated Projection-Reconstruction MRcited by 31
- 2021An End-To-End-Trainable Iterative Network Architecture for Accelerated Radial Multi-Coil 2D Cine MR Image Reconstructioncited by 7
- 2023Bicubic++: Slim, Slimmer, Slimmest -- Designing an Industry-Grade Super-Resolution Networkcited by 6
3 more
- 2021Memorization in Deep Neural Networks: Does the Loss Function matter?cited by 5
- 2022Dynamic Cardiac MRI Reconstruction Using Combined Tensor Nuclear Norm and Casorati Matrix Nuclear Norm Regularizationscited by 2
- 2023Reconstruction of Cardiac Cine MRI Using Motion-Guided Deformable Alignment and Multi-Resolution Fusioncited by 1
Built on by
Citing papers per year: 2024: 1.
1 paper.