Super-resolution in MRI: better images faster?

Plenge, Esben; Poot, Dirk H. J.; Bernsen, Monique; Kotek, Gyula; Houston, Gavin C.; Wielopolski, Piotr A.; Weerd, Louise van der; Niessen, Wiro J.; Meijering, Erik

doi:10.1117/12.911235

Cited by 9 publications

(11 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The constrained least square regularization method uses smoothness, and regularized Tikhonov leastsquare estimator uses l 2 -norm as regularization. 27 The l 2 -norm does not guarantee a unique solution. Farsiu et al 28 exploited an alternative l 1 -norm minimization for fast and robust SR. Zomet and colleagues 29 described a robust SR method for considering the outliers.…”

Section: Regularization Approachesmentioning

confidence: 99%

Comparison of super-resolution algorithms applied to retinal images

et al. 2014

View full text Add to dashboard Cite

Abstract. A critical challenge in biomedical imaging is to optimally balance the trade-off among image resolution, signal-to-noise ratio, and acquisition time. Acquiring a high-resolution image is possible; however, it is either expensive or time consuming or both. Resolution is also limited by the physical properties of the imaging device, such as the nature and size of the input source radiation and the optics of the device. Super-resolution (SR), which is an off-line approach for improving the resolution of an image, is free of these trade-offs. Several methodologies, such as interpolation, frequency domain, regularization, and learning-based approaches, have been developed over the past several years for SR of natural images. We review some of these methods and demonstrate the positive impact expected from SR of retinal images and investigate the performance of various SR techniques. We use a fundus image as an example for simulations.

show abstract

Section: Regularization Approachesmentioning

confidence: 99%

Comparison of super-resolution algorithms applied to retinal images

et al. 2014

View full text Add to dashboard Cite

show abstract

“…(2007) shows that using SR techniques produces better contrast ratios and better target-to-background ratios than the standard reconstructions. Plenge et al. (2012) designed, instead, an experimental framework to show that the SR reconstructions are more advantageous in terms of the SNR with respect to the direct HR acquisition.…”

Section: Discussionmentioning

confidence: 99%

Adversarial training with cycle consistency for unsupervised super-resolution in endomicroscopy

Ravì

Szczotka

Pereira

et al. 2019

Medical Image Analysis

View full text Add to dashboard Cite

HighlightsWe propose an unsupervised SR framework based on an adversarial deep neural network with a physically-inspired cycle consistency.Our framework can exploit HR images, regardless of the domain where they are coming from, to transfer the quality of the HR images to the initial LR images.Our system can be particularly useful in all situations where pairs of LR/HR are not available during the training of a super-resolution algorithm.Our quantitative analysis, validated using a database of 238 endomicroscopy video sequences from 143 patients, shows the ability of the pipeline to produce convincing super-resolved images.A Mean Opinion Score (MOS) study also confirms this quantitative image quality assessment.

show abstract

“…Early research on super-resolution MRI images used super-resolution reconstruction (SRR) to improve the image resolution. SRR combines a series of low-resolution MRI images into a high-resolution image [ 5 , 6 , 7 ]. This method requires large time and equipment costs, and subsequent research has shown that adding more low-resolution scans does not necessarily improve the resolution [ 8 , 9 ].…”

Section: Introductionmentioning

confidence: 99%

“…Initially, SISR used a form of regularization conditions and then used prior knowledge to enhance the reconstruction ability of linear models [ 11 ]. However, this type of method is computationally complex and requires many computing resources [ 5 , 7 , 12 ].…”

Section: Introductionmentioning

confidence: 99%

“…Based on the convolutional neural network (CNN) model, the super-resolution CNN (SRCNN) [ 13 ] was introduced into SISR. MRI super-resolution research attempted to combine convolutional neural networks for super-resolution reconstruction [ 5 , 7 , 8 , 9 ]. With the introduction of residual networks [ 14 , 15 , 16 , 17 , 18 , 19 ], super-resolution technology based on deep learning began to develop into deeper networks [ 14 , 15 , 16 ].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

3D MRI Reconstruction Based on 2D Generative Adversarial Network Super-Resolution

Zhang

Shinomiya

Yoshida

2021

Sensors

View full text Add to dashboard Cite

The diagnosis of brain pathologies usually involves imaging to analyze the condition of the brain. Magnetic resonance imaging (MRI) technology is widely used in brain disorder diagnosis. The image quality of MRI depends on the magnetostatic field strength and scanning time. Scanners with lower field strengths have the disadvantages of a low resolution and high imaging cost, and scanning takes a long time. The traditional super-resolution reconstruction method based on MRI generally states an optimization problem in terms of prior information. It solves the problem using an iterative approach with a large time cost. Many methods based on deep learning have emerged to replace traditional methods. MRI super-resolution technology based on deep learning can effectively improve MRI resolution through a three-dimensional convolutional neural network; however, the training costs are relatively high. In this paper, we propose the use of two-dimensional super-resolution technology for the super-resolution reconstruction of MRI images. In the first reconstruction, we choose a scale factor of 2 and simulate half the volume of MRI slices as input. We utilize a receiving field block enhanced super-resolution generative adversarial network (RFB-ESRGAN), which is superior to other super-resolution technologies in terms of texture and frequency information. We then rebuild the super-resolution reconstructed slices in the MRI. In the second reconstruction, the image after the first reconstruction is composed of only half of the slices, and there are still missing values. In our previous work, we adopted the traditional interpolation method, and there was still a gap in the visual effect of the reconstructed images. Therefore, we propose a noise-based super-resolution network (nESRGAN). The noise addition to the network can provide additional texture restoration possibilities. We use nESRGAN to further restore MRI resolution and high-frequency information. Finally, we achieve the 3D reconstruction of brain MRI images through two super-resolution reconstructions. Our proposed method is superior to 3D super-resolution technology based on deep learning in terms of perception range and image quality evaluation standards.

show abstract

Super-resolution in MRI: better images faster?

Cited by 9 publications

References 8 publications

Comparison of super-resolution algorithms applied to retinal images

Comparison of super-resolution algorithms applied to retinal images

Adversarial training with cycle consistency for unsupervised super-resolution in endomicroscopy

3D MRI Reconstruction Based on 2D Generative Adversarial Network Super-Resolution

Contact Info

Product

Resources

About