{"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/anatomical-priors-in-convolutional-networks-1","title":"Anatomical Priors in Convolutional Networks for Unsupervised Biomedical Segmentation","arxiv_id":"1903.03148","date":"2019-03-07","proceeding":"CVPR 2018 6","authors":["Adrian V. Dalca","John Guttag","Mert R. Sabuncu"],"abstract":"We consider the problem of segmenting a biomedical image into anatomical\nregions of interest. We specifically address the frequent scenario where we\nhave no paired training data that contains images and their manual\nsegmentations. Instead, we employ unpaired segmentation images to build an\nanatomical prior. Critically these segmentations can be derived from imaging\ndata from a different dataset and imaging modality than the current task. We\nintroduce a generative probabilistic model that employs the learned prior\nthrough a convolutional neural network to compute segmentations in an\nunsupervised setting. We conducted an empirical analysis of the proposed\napproach in the context of structural brain MRI segmentation, using a\nmulti-study dataset of more than 14,000 scans. Our results show that an\nanatomical prior can enable fast unsupervised segmentation which is typically\nnot possible using standard convolutional networks. The integration of\nanatomical priors can facilitate CNN-based anatomical segmentation in a range\nof novel clinical problems, where few or no annotations are available and thus\nstandard networks are not trainable. The code is freely available at\nhttp://github.com/adalca/neuron.","url_abs":"http://arxiv.org/abs/1903.03148v1","url_pdf":"http://arxiv.org/pdf/1903.03148v1.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":"anatomical-priors-in-convolutional-networks-1","repo_url":"https://github.com/adalca/neuron","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"anatomical-priors-in-convolutional-networks-1","repo_url":"https://github.com/adalca/neurite","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"mri-segmentation","task_name":"MRI segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1903.03148","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}