Figure - available from: Journal of Clinical Medicine
This content is subject to copyright.
Representative case of lesion conspicuity on DL-FLAIR images. Focal marginal gliosis showing hyperintensity (arrows) is seen at the anterior aspect of the surgical cavity in the right cerebellum (a–c). The lesion is well delineated on the native synthetic FLAIR (a), DL-FLAIR (b), and conventional FLAIR images (c). However, the hyperintense lesion on the native synthetic FLAIR (arrow on a) is incompletely preserved on DL-FLAIR image (arrow on b), showing a decrease in the degree of its hyperintensity. Flow artifacts in the fourth ventricle on the native synthetic FLAIR (arrowhead on a) are successfully removed on DL-FLAIR image (arrowhead on b). Multiple FLAIR hyperintense lesions in both centrum semiovale, suggesting grade II small vessel disease on the native synthetic FLAIR (d) are well preserved on DL-FLAIR image (e). (f) Conventional FLAIR image is shown for comparison.
Source publication
We investigated the capability of a trained deep learning (DL) model with a convolutional neural network (CNN) in a different scanning environment in terms of ameliorating the quality of synthetic fluid-attenuated inversion recovery (FLAIR) images. The acquired data of 319 patients obtained from the retrospective review were used as test sets for t...
Citations
... More recently, machine learning and deep learning methods have attracted an increasing interest in the areas of MRI quality control (QC) and quality assurance (QA) processes [15][16][17][18][19][20][21][22][23][24][25][26][27] and implementations of retrospective artifact reduction [15,16,20]- [25,28]. Deep learningbased approaches have been used for motion detection [15], evaluation of image quality [21,22], and furthermore, as a decision-making tool to decide if an imaging series is clinically useful and therefore decrease rescan and recall rates [25]. ...
... In addition, a separate diagnostic comparison study performed by the neuroradiologist demonstrated that none of the abnormal findings and/ or subtle structural changes on the originally acquired images was missing out on the DRN-DCMB images. Overall, the DRN-DCMB model effectively removed motion artifacts while maintained the image contrast, reserved the conspicuity of the small anatomic structures, and precluded the appearance of enhancing image edges, which were observed in images generated by other CNN networks [18,19,24,28]. ...
Recently, artificial intelligence (AI) technology has shown potential clinical utility in a wide range of MRI fields. In particular, AI models for improving the efficiency of the image acquisition process and the quality of reconstructed images are being actively developed by the MR research community. AI is expected to further reduce acquisition times in various MRI protocols used in clinical practice when compared to current parallel imaging techniques. Additionally, AI can help with tasks such as planning, parameter optimization, artifact reduction, and quality assessment. Furthermore, AI is being actively applied to automate MR image analysis such as image registration, segmentation, and object detection. For this reason, it is important to consider the effects of protocols or devices in MR image analysis. In this review article, we briefly introduced issues related to AI application of MR image acquisition and reconstruction.
Synthetic MRI is a technique that synthesizes contrast‐weighted images from multicontrast MRI data. There have been advances in synthetic MRI since the technique was introduced. Although a number of synthetic MRI methods have been developed for quantifying one or more relaxometric parameters and for generating multiple contrast‐weighted images, this review focuses on several methods that quantify all three relaxometric parameters (T1, T2, and proton density) and produce multiple contrast‐weighted images. Acquisition, quantification, and image synthesis techniques are discussed for each method. We discuss the image quality and diagnostic accuracy of synthetic MRI methods and their clinical applications in neuroradiology. Based on this analysis, we highlight areas that need to be addressed for synthetic MRI to be widely implemented in the clinic.
Level of Evidence
5
Technical Efficacy Stage
1
