{"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/label-driven-weakly-supervised-learning-for","title":"Label-driven weakly-supervised learning for multimodal deformable image registration","arxiv_id":"1711.01666","date":"2017-11-05","proceeding":null,"authors":["Yipeng Hu","Marc Modat","Eli Gibson","Nooshin Ghavami","Ester Bonmati","Caroline M. Moore","Mark Emberton","J. Alison Noble","Dean C. Barratt","Tom Vercauteren"],"abstract":"Spatially aligning medical images from different modalities remains a\nchallenging task, especially for intraoperative applications that require fast\nand robust algorithms. We propose a weakly-supervised, label-driven formulation\nfor learning 3D voxel correspondence from higher-level label correspondence,\nthereby bypassing classical intensity-based image similarity measures. During\ntraining, a convolutional neural network is optimised by outputting a dense\ndisplacement field (DDF) that warps a set of available anatomical labels from\nthe moving image to match their corresponding counterparts in the fixed image.\nThese label pairs, including solid organs, ducts, vessels, point landmarks and\nother ad hoc structures, are only required at training time and can be\nspatially aligned by minimising a cross-entropy function of the warped moving\nlabel and the fixed label. During inference, the trained network takes a new\nimage pair to predict an optimal DDF, resulting in a fully-automatic,\nlabel-free, real-time and deformable registration. For interventional\napplications where large global transformation prevails, we also propose a\nneural network architecture to jointly optimise the global- and local\ndisplacements. Experiment results are presented based on cross-validating\nregistrations of 111 pairs of T2-weighted magnetic resonance images and 3D\ntransrectal ultrasound images from prostate cancer patients with a total of\nover 4000 anatomical labels, yielding a median target registration error of 4.2\nmm on landmark centroids and a median Dice of 0.88 on prostate glands.","url_abs":"http://arxiv.org/abs/1711.01666v2","url_pdf":"http://arxiv.org/pdf/1711.01666v2.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":"label-driven-weakly-supervised-learning-for","repo_url":"https://github.com/yipenghu/label-reg","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"image-registration","task_name":"Image Registration"},{"task_slug":"weakly-supervised-learning","task_name":"Weakly-supervised Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.01666","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}