{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/efficient-and-accurate-mri-super-resolution","title":"Efficient and Accurate MRI Super-Resolution using a Generative Adversarial Network and 3D Multi-Level Densely Connected Network","arxiv_id":"1803.01417","date":"2018-03-04","proceeding":null,"authors":["Yuhua Chen","Feng Shi","Anthony G. Christodoulou","Zhengwei Zhou","Yibin Xie","Debiao Li"],"abstract":"High-resolution (HR) magnetic resonance images (MRI) provide detailed\nanatomical information important for clinical application and quantitative\nimage analysis. However, HR MRI conventionally comes at the cost of longer scan\ntime, smaller spatial coverage, and lower signal-to-noise ratio (SNR). Recent\nstudies have shown that single image super-resolution (SISR), a technique to\nrecover HR details from one single low-resolution (LR) input image, could\nprovide high-quality image details with the help of advanced deep convolutional\nneural networks (CNN). However, deep neural networks consume memory heavily and\nrun slowly, especially in 3D settings. In this paper, we propose a novel 3D\nneural network design, namely a multi-level densely connected super-resolution\nnetwork (mDCSRN) with generative adversarial network (GAN)-guided training. The\nmDCSRN quickly trains and inferences and the GAN promotes realistic output\nhardly distinguishable from original HR images. Our results from experiments on\na dataset with 1,113 subjects show that our new architecture beats other\npopular deep learning methods in recovering 4x resolution-downgraded im-ages\nand runs 6x faster.","url_abs":"http://arxiv.org/abs/1803.01417v3","url_pdf":"http://arxiv.org/pdf/1803.01417v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"efficient-and-accurate-mri-super-resolution","repo_url":"https://github.com/Hadrien-Cornier/E6040-super-resolution-project","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"efficient-and-accurate-mri-super-resolution","repo_url":"https://github.com/seobeomjin/ML_Project_MRI_Brain_Image_SR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":null,"task_name":"Generative Adversarial Network"},{"task_slug":"image-super-resolution","task_name":"Image Super-Resolution"},{"task_slug":"super-resolution","task_name":"Super-Resolution"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1803.01417","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}