{"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/towards-automatic-initialization-of","title":"Towards automatic initialization of registration algorithms using simulated endoscopy images","arxiv_id":"1806.10748","date":"2018-06-28","proceeding":null,"authors":["Ayushi Sinha","Masaru Ishii","Russell H. Taylor","Gregory D. Hager","Austin Reiter"],"abstract":"Registering images from different modalities is an active area of research in\ncomputer aided medical interventions. Several registration algorithms have been\ndeveloped, many of which achieve high accuracy. However, these results are\ndependent on many factors, including the quality of the extracted features or\nsegmentations being registered as well as the initial alignment. Although\nseveral methods have been developed towards improving segmentation algorithms\nand automating the segmentation process, few automatic initialization\nalgorithms have been explored. In many cases, the initial alignment from which\na registration is initiated is performed manually, which interferes with the\nclinical workflow. Our aim is to use scene classification in endoscopic\nprocedures to achieve coarse alignment of the endoscope and a preoperative\nimage of the anatomy. In this paper, we show using simulated scenes that a\nneural network can predict the region of anatomy (with respect to a\npreoperative image) that the endoscope is located in by observing a single\nendoscopic video frame. With limited training and without any hyperparameter\ntuning, our method achieves an accuracy of 76.53 (+/-1.19)%. There are several\navenues for improvement, making this a promising direction of research. Code is\navailable at https://github.com/AyushiSinha/AutoInitialization.","url_abs":"http://arxiv.org/abs/1806.10748v1","url_pdf":"http://arxiv.org/pdf/1806.10748v1.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":"towards-automatic-initialization-of","repo_url":"https://github.com/AyushiSinha/AutoInitialization","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"anatomy","task_name":"Anatomy"},{"task_slug":"scene-classification","task_name":"Scene Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}