{"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/predicting-slice-to-volume-transformation-in","title":"Predicting Slice-to-Volume Transformation in Presence of Arbitrary Subject Motion","arxiv_id":"1702.08891","date":"2017-02-28","proceeding":null,"authors":["Benjamin Hou","Amir Alansary","Steven McDonagh","Alice Davidson","Mary Rutherford","Jo V. Hajnal","Daniel Rueckert","Ben Glocker","Bernhard Kainz"],"abstract":"This paper aims to solve a fundamental problem in intensity-based 2D/3D\nregistration, which concerns the limited capture range and need for very good\ninitialization of state-of-the-art image registration methods. We propose a\nregression approach that learns to predict rotation and translations of\narbitrary 2D image slices from 3D volumes, with respect to a learned canonical\natlas co-ordinate system. To this end, we utilize Convolutional Neural Networks\n(CNNs) to learn the highly complex regression function that maps 2D image\nslices into their correct position and orientation in 3D space. Our approach is\nattractive in challenging imaging scenarios, where significant subject motion\ncomplicates reconstruction performance of 3D volumes from 2D slice data. We\nextensively evaluate the effectiveness of our approach quantitatively on\nsimulated MRI brain data with extreme random motion. We further demonstrate\nqualitative results on fetal MRI where our method is integrated into a full\nreconstruction and motion compensation pipeline. With our CNN regression\napproach we obtain an average prediction error of 7mm on simulated data, and\nconvincing reconstruction quality of images of very young fetuses where\nprevious methods fail. We further discuss applications to Computed Tomography\nand X-ray projections. Our approach is a general solution to the 2D/3D\ninitialization problem. It is computationally efficient, with prediction times\nper slice of a few milliseconds, making it suitable for real-time scenarios.","url_abs":"http://arxiv.org/abs/1702.08891v2","url_pdf":"http://arxiv.org/pdf/1702.08891v2.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":"predicting-slice-to-volume-transformation-in","repo_url":"https://github.com/farrell236/SVRnet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"image-registration","task_name":"Image Registration"},{"task_slug":"motion-compensation","task_name":"Motion Compensation"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}