Corpus ID: 233481868

Simultaneous super-resolution and motion artifact removal in diffusion-weighted MRI using unsupervised deep learning

@article{Chung2021SimultaneousSA,
  title={Simultaneous super-resolution and motion artifact removal in diffusion-weighted MRI using unsupervised deep learning},
  author={Hyungjin Chung and Jaehyun Kim and Jeong Hee Yoon and Jeong Min Lee and Jong-Chul Ye},
  journal={ArXiv},
  year={2021},
  volume={abs/2105.00240}
}
Diffusion-weighted MRI is nowadays performed routinely due to its prognostic ability, yet the quality of the scans are often unsatisfactory which can subsequently hamper the clinical utility. To overcome the limitations, here we propose a fully unsupervised quality enhancement scheme, which boosts the resolution and removes the motion artifact simultaneously. This process is done by first training the network using optimal transport driven cycleGAN with stochastic degradation block which learns… Expand
1 Citations
3D High-Quality Magnetic Resonance Image Restoration in Clinics Using Deep Learning
  • Hao Li, Jianan Liu
  • Engineering, Computer Science
  • 2021
Shortening acquisition time and reducing the motion-artifact are two of the most essential concerns in magnetic resonance imaging. As a promising solution, deep learning-based high-quality MR imageExpand

References

SHOWING 1-10 OF 40 REFERENCES
ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks
TLDR
This work thoroughly study three key components of SRGAN – network architecture, adversarial loss and perceptual loss, and improves each of them to derive an Enhanced SRGAN (ESRGAN), which achieves consistently better visual quality with more realistic and natural textures than SRGAN. Expand
Diffusion weighted imaging: Technique and applications
TLDR
This review article provides insights in to the evolution of DWI as a new imaging paradigm and provides a summary of current role ofDWI in various disease processes. Expand
Two-Stage Deep Learning for Accelerated 3D Time-of-Flight MRA without Matched Training Data
TLDR
Extensive experiments demonstrate that the proposed learning process without matched reference exceeds performance of state-of-the-art compressed sensing (CS)-based method and provides comparable or even better results than supervised learning approaches. Expand
Unpaired Training of Deep Learning tMRA for Flexible Spatio-Temporal Resolution
TLDR
A novel unpaired training scheme for deep learning using optimal transport driven cycle-consistent generative adversarial network (cycleGAN) with just a single pair of generator and discriminator is proposed, which makes the training much simpler but still improves the performance. Expand
Geometric Approaches to Increase the Expressivity of Deep Neural Networks for MR Reconstruction
TLDR
A systematic geometric approach using bootstrapping and subnetwork aggregation using an attention module to increase the expressivity of the underlying neural network to improve reconstruction performance with negligible complexity increases is proposed. Expand
Unpaired Deep Learning for Accelerated MRI Using Optimal Transport Driven CycleGAN
TLDR
An unpaired deep learning approach using a optimal transport driven cycle-consistent generative adversarial network (OT-cycleGAN) that employs a single pair of generator, and discriminator that is rigorously derived from a dual formulation of the optimal transport formulation using a specially designed penalized least squares cost. Expand
Unsupervised MR Motion Artifact Deep Learning using Outlier-Rejecting Bootstrap Aggregation
TLDR
A novel unsupervised deep learning scheme through outlier-rejecting bootstrap subsampling and aggregation that can be applied for artifact correction from simulated motion as well as real motion from TSM successfully, outperforming existing state-of-the-art deep learning methods. Expand
U-Net: Convolutional Networks for Biomedical Image Segmentation
TLDR
It is shown that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Expand
Diffusion‐weighted imaging of the abdomen at 3.0 Tesla: Image quality and apparent diffusion coefficient reproducibility compared with 1.5 Tesla
To compare single‐shot echo‐planar imaging (SS EPI) diffusion‐weighted MRI (DWI) of abdominal organs between 1.5 Tesla (T) and 3.0T in healthy volunteers in terms of image quality, apparent diffusionExpand
...
1
2
3
4
...