Purpose Gadolinium‐based contrast agents (GBCAs) have been successfully applied in magnetic resonance (MR) imaging to facilitate better lesion visualization. However, gadolinium deposition in the human brain raised widespread concerns recently. On the other hand, although high‐resolution three‐dimensional (3D) MR images are more desired for most existing medical image processing algorithms, their long scan duration and high acquiring costs make 2D MR images still much more common clinically. Therefore, developing alternative solutions for 3D contrast‐enhanced MR image synthesis to replace GBCAs injection becomes an urgent requirement. Methods This study proposed a deep learning framework that produces 3D isotropic full‐contrast T2Flair images from 2D anisotropic noncontrast T2Flair image stacks. The super‐resolution (SR) and contrast‐enhanced (CE) synthesis tasks are completed in sequence by using an identical generative adversarial network (GAN) with the same techniques. To solve the problem that intramodality datasets from different scanners have specific combinations of orientations, contrasts, and resolutions, we conducted a region‐based data augmentation technique on the fly during training to simulate various imaging protocols in the clinic. We further improved our network by introducing atrous spatial pyramid pooling, enhanced residual blocks, and deep supervision for better quantitative and qualitative results. Results Our proposed method achieved superior CE‐synthesized performance in quantitative metrics and perceptual evaluation. In detail, the PSNR, structural‐similarity‐index, and AUC are 32.25 dB, 0.932, and 0.991 in the whole brain and 24.93 dB, 0.851, and 0.929 in tumor regions. The radiologists’ evaluations confirmed that our proposed method has high confidence in the diagnosis. Analysis of the generalization ability showed that benefiting from the proposed data augmentation technique, our network can be applied to “unseen” datasets with slight drops in quantitative and qualitative results. Conclusion Our work demonstrates the clinical potential of synthesizing diagnostic 3D isotropic CE brain MR images from a single 2D anisotropic noncontrast sequence.
Background Hippocampal alterations have been implicated in the pathophysiology of cognitive impairment in hemodialysis patients. The hippocampus consists of several distinct subfields, and the molecular mechanisms underlying cognition might be associated with specific hippocampal subfield volume changes. However, this has not yet been investigated in hemodialysis patients. This study aimed to explore volumetric abnormalities in hippocampal subfields in regular hemodialysis patients. Methods High-resolution T1-weighted structural images were collected in 61 subjects including 36 hemodialysis patients and 25 healthy controls. A state-of-the-art hippocampal segmentation approach was adopted to segment the hippocampal subfields. Group differences in hippocampal subfield volumes were assessed in Python with a statsmodels module using an ordinary least squares regression with age and sex as nuisance effects. Results Hemodialysis patients had significantly smaller volumes in the bilateral hippocampus (p < 0.05/2, Bonferroni corrected), CA (cornu ammonis) 1, CA4, granule cell and molecular layer of the dentate gyrus, hippocampus-amygdala-transition-area and molecular layer of the hippocampus than healthy controls (p < 0.05/24, Bonferroni corrected). Hemodialysis patients also had lower volumes in the left hippocampal tail and right fimbria than healthy controls (p < 0.05/24, Bonferroni corrected). Hippocampal subfield volumes were associated with neuropsychological test scores, the duration of disease and hemoglobin levels. Conclusions We found smaller hippocampal subfield volumes in hemodialysis patients, which were associated with impaired cognition, supporting their role in memory disturbance in the hemodialysis population. However, multiple clinical factors may have confounded the results, and therefore, the interpretation of these results needs to be cautious.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.