Deep Learning for Image Enhancement and Correction in Magnetic Resonance Imaging—State-of-the-Art and Challenges

Abstract

Magnetic resonance imaging (MRI) provides excellent soft-tissue contrast for clinical diagnoses and research which underpin many recent breakthroughs in medicine and biology. The post-processing of reconstructed MR images is often automated for incorporation into MRI scanners by the manufacturers and increasingly plays a critical role in the final image quality for clinical reporting and interpretation. For image enhancement and correction, the post-processing steps include noise reduction, image artefact correction, and image resolution improvements. With the recent success of deep learning in many research fields, there is great potential to apply deep learning for MR image enhancement, and recent publications have demonstrated promising results. Motivated by the rapidly growing literature in this area, in this review paper, we provide a comprehensive overview of deep learning-based methods for post-processing MR images to enhance image quality and correct image artefacts. We aim to provide researchers in MRI or other research fields, including computer vision and image processing, a literature survey of deep learning approaches for MR image enhancement. We discuss the current limitations of the application of artificial intelligence in MRI and highlight possible directions for future developments. In the era of deep learning, we highlight the importance of a critical appraisal of the explanatory information provided and the generalizability of deep learning algorithms in medical imaging.

AdvancedMRTechniques for Preoperative Glioma Characterization: Part 2

“…8). While AI‐driven methods can also be used for MRI acquisition, postprocessing, and analysis in general, this lies beyond the scope of this review and is covered by elsewhere 98,99 …”

“…While AI-driven methods can also be used for MRI acquisition, postprocessing, and analysis in general, this lies beyond the scope of this review and is covered by elsewhere. 98,99 CLINICAL APPLICATION. Most of the radiomics studies focus on working with the conventional structural and diffusion MRI data as these are acquired in clinical routine, allowing easy translation and securing large datasets needed for training and validation.…”

The effect of head motion on brain age prediction using deep convolutional neural networks

“…In addition, several methods are available for the correction of motion-corrupted images-e.g. prospective motion correction using volumetric navigators (vNavs) can be applied effectively to reduce the motion-induced bias in morphometric estimates [85], and recent deep learning-based algorithms have provided promising results in the retrospective removal of motion artifacts from MRI scans [86,87], although the potential benefit of these algorithms in the context of brain age prediction remains to be investigated.…”

DeepCor: Denoising fMRI Data with Contrastive Autoencoders

“…To overcome this limitation, the field is shifting towards denoising approaches based on deep learning ([9, 10, 11, 12]). While initial results are promising, some of these approaches are only applicable to task-based experiments [9], and others require large numbers of participants [11].…”