{"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/3d-fully-convolutional-networks-for","title":"3D fully convolutional networks for subcortical segmentation in MRI: A large-scale study","arxiv_id":"1612.03925","date":"2016-12-12","proceeding":null,"authors":["J. Dolz","C. Desrosiers","I. Ben Ayed"],"abstract":"This study investigates a 3D and fully convolutional neural network (CNN) for\nsubcortical brain structure segmentation in MRI. 3D CNN architectures have been\ngenerally avoided due to their computational and memory requirements during\ninference. We address the problem via small kernels, allowing deeper\narchitectures. We further model both local and global context by embedding\nintermediate-layer outputs in the final prediction, which encourages\nconsistency between features extracted at different scales and embeds\nfine-grained information directly in the segmentation process. Our model is\nefficiently trained end-to-end on a graphics processing unit (GPU), in a single\nstage, exploiting the dense inference capabilities of fully CNNs.\n  We performed comprehensive experiments over two publicly available datasets.\nFirst, we demonstrate a state-of-the-art performance on the ISBR dataset. Then,\nwe report a {\\em large-scale} multi-site evaluation over 1112 unregistered\nsubject datasets acquired from 17 different sites (ABIDE dataset), with ages\nranging from 7 to 64 years, showing that our method is robust to various\nacquisition protocols, demographics and clinical factors. Our method yielded\nsegmentations that are highly consistent with a standard atlas-based approach,\nwhile running in a fraction of the time needed by atlas-based methods and\navoiding registration/normalization steps. This makes it convenient for massive\nmulti-site neuroanatomical imaging studies. To the best of our knowledge, our\nwork is the first to study subcortical structure segmentation on such\nlarge-scale and heterogeneous data.","url_abs":"http://arxiv.org/abs/1612.03925v2","url_pdf":"http://arxiv.org/pdf/1612.03925v2.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":"3d-fully-convolutional-networks-for","repo_url":"https://github.com/josedolz/3D-F-CNN-BrainStruct","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null},{"paper_slug":"3d-fully-convolutional-networks-for","repo_url":"https://github.com/josedolz/LiviaNET","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"3d-fully-convolutional-networks-for","repo_url":"https://github.com/josedolz/LiviaNet_pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"3d-medical-imaging-segmentation","task_name":"3D Medical Imaging Segmentation"},{"task_slug":"brain-segmentation","task_name":"Brain Segmentation"},{"task_slug":null,"task_name":"GPU"},{"task_slug":"medical-image-segmentation","task_name":"Medical Image Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}