{"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/hemis-hetero-modal-image-segmentation","title":"HeMIS: Hetero-Modal Image Segmentation","arxiv_id":"1607.05194","date":"2016-07-18","proceeding":null,"authors":["Mohammad Havaei","Nicolas Guizard","Nicolas Chapados","Yoshua Bengio"],"abstract":"We introduce a deep learning image segmentation framework that is extremely\nrobust to missing imaging modalities. Instead of attempting to impute or\nsynthesize missing data, the proposed approach learns, for each modality, an\nembedding of the input image into a single latent vector space for which\narithmetic operations (such as taking the mean) are well defined. Points in\nthat space, which are averaged over modalities available at inference time, can\nthen be further processed to yield the desired segmentation. As such, any\ncombinatorial subset of available modalities can be provided as input, without\nhaving to learn a combinatorial number of imputation models. Evaluated on two\nneurological MRI datasets (brain tumors and MS lesions), the approach yields\nstate-of-the-art segmentation results when provided with all modalities;\nmoreover, its performance degrades remarkably gracefully when modalities are\nremoved, significantly more so than alternative mean-filling or other synthesis\napproaches.","url_abs":"http://arxiv.org/abs/1607.05194v1","url_pdf":"http://arxiv.org/pdf/1607.05194v1.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":"hemis-hetero-modal-image-segmentation","repo_url":"https://github.com/momih/pc-hemis","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-segmentation","task_name":"Image Segmentation"},{"task_slug":"imputation","task_name":"Imputation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semantic-segmentation-on-nyu-depth-v2","task":"Semantic Segmentation","dataset":"NYU Depth v2","model":"HeMIS","rank_in_archive_order":112,"of":121,"metrics":{"Mean IoU":"37.77%"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1607.05194","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}