Deep learning in radiology: An overview of the concepts and a survey of the state of the art with focus on MRI

Mazurowski, Maciej A.; Buda, Mateusz; Saha, Ashirbani; Bashir, Mustafa R.

doi:10.1002/jmri.26534

Cited by 398 publications

(247 citation statements)

References 147 publications

Supporting

Mentioning

218

Contrasting

Unclassified

Order By: Relevance

“…Deep neural networks (DNNs) have been successfully applied to various image processing tasks . In particular, the generative adversarial network (GAN) has been widely used for image‐to‐image translation .…”

Section: Introductionmentioning

confidence: 99%

Visual enhancement of Cone‐beam CT by use of CycleGAN

et al. 2020

View full text Add to dashboard Cite

Purpose Cone‐beam computed tomography (CBCT) offers advantages over conventional fan‐beam CT in that it requires a shorter time and less exposure to obtain images. However, CBCT images suffer from low soft‐tissue contrast, noise, and artifacts compared to conventional fan‐beam CT images. Therefore, it is essential to improve the image quality of CBCT. Methods In this paper, we propose a synthetic approach to translate CBCT images with deep neural networks. Our method requires only unpaired and unaligned CBCT images and planning fan‐beam CT (PlanCT) images for training. The CBCT images and PlanCT images may be obtained from other patients as long as they are acquired with the same scanner settings. Once trained, three‐dimensionally reconstructed CBCT images can be directly translated into high‐quality PlanCT‐like images. Results We demonstrate the effectiveness of our method with images obtained from 20 prostate patients, and provide a statistical and visual comparison. The image quality of the translated images shows substantial improvement in voxel values, spatial uniformity, and artifact suppression compared to those of the original CBCT. The anatomical structures of the original CBCT images were also well preserved in the translated images. Conclusions Our method produces visually PlanCT‐like images from CBCT images while preserving anatomical structures.

show abstract

Section: Introductionmentioning

confidence: 99%

Visual enhancement of Cone‐beam CT by use of CycleGAN

et al. 2020

View full text Add to dashboard Cite

show abstract

“…In recent years, machine learning and deep learning have become increasingly popular topics for research, and applications in MRI are frequent . Machine and deep learning may be a natural fit to solve some of the challenges in MRF, such as image reconstruction and pattern matching.…”

Section: Applications Of Machine and Deep Learning To Mrfmentioning

confidence: 99%

Magnetic resonance fingerprinting review part 2: Technique and directions

McGivney

Boyacıoğlu

Jiang

et al. 2019

Magnetic Resonance Imaging

View full text Add to dashboard Cite

Magnetic resonance fingerprinting (MRF) is a general framework to quantify multiple MR‐sensitive tissue properties with a single acquisition. There have been numerous advances in MRF in the years since its inception. In this work we highlight some of the recent technical developments in MRF, focusing on sequence optimization, modifications for reconstruction and pattern matching, new methods for partial volume analysis, and applications of machine and deep learning. Level of Evidence: 2 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2020;51:993–1007.

show abstract

“…Deep‐learning frameworks are able to "learn" standard MRI reconstruction techniques, such as Cartesian and non‐Cartesian acquisition schemes. Combining deep learning with k ‐space undersampling with model‐based/compressed sensing reconstruction schemes has the potential to revolutionize imaging science by optimizing the methods image data collected . A recent report from Lee et al described that a proposed deep‐learning framework may have great potential for accelerated MR reconstruction by generating accurate results immediately.…”

Section: Techniques To Minimize Sedation For Pediatric Mri Examinationsmentioning

confidence: 99%

Techniques for minimizing sedation in pediatric MRI

Dong

Zhu

Bulas

2019

Magnetic Resonance Imaging

View full text Add to dashboard Cite

MRI is used widely in infants and young children. However, in these young cases deep sedation or general anesthesia is often required to minimize motion artifacts during MRI examinations. Although the benefits of MR typically outweigh the potential risks of sedation when delivered by an experienced team, there are increasing concerns regarding the affect of sedation on young children. There continues to be a push to develop various strategies that can minimize the need for sedation. The present review summarizes several technical and clinical approaches that can help decrease the need for sedation in the pediatric patient. Optimization of the MRI environment, the role of child life specialists, feed‐and‐bundle and distraction techniques, noise‐reduction methods, artificial intelligence, and MRI advances to decrease both scan times and motion artifacts will be discussed. Level of Evidence: 5 Technical Efficacy Stage: 1 J. Magn. Reson. Imaging 2019.

show abstract

Deep learning in radiology: An overview of the concepts and a survey of the state of the art with focus on MRI

Abstract: View this article online at wileyonlinelibrary.com.

Cited by 398 publications

References 147 publications

Visual enhancement of Cone‐beam CT by use of CycleGAN

Visual enhancement of Cone‐beam CT by use of CycleGAN

Magnetic resonance fingerprinting review part 2: Technique and directions

Techniques for minimizing sedation in pediatric MRI

Contact Info

Product

Resources

About