{"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/ivus-net-an-intravascular-ultrasound","title":"IVUS-Net: An Intravascular Ultrasound Segmentation Network","arxiv_id":"1806.03583","date":"2018-06-10","proceeding":null,"authors":["Ji Yang","Lin Tong","Mehdi Faraji","Anup Basu"],"abstract":"IntraVascular UltraSound (IVUS) is one of the most effective imaging\nmodalities that provides assistance to experts in order to diagnose and treat\ncardiovascular diseases. We address a central problem in IVUS image analysis\nwith Fully Convolutional Network (FCN): automatically delineate the lumen and\nmedia-adventitia borders in IVUS images, which is crucial to shorten the\ndiagnosis process or benefits a faster and more accurate 3D reconstruction of\nthe artery. Particularly, we propose an FCN architecture, called IVUS-Net,\nfollowed by a post-processing contour extraction step, in order to\nautomatically segments the interior (lumen) and exterior (media-adventitia)\nregions of the human arteries. We evaluated our IVUS-Net on the test set of a\nstandard publicly available dataset containing 326 IVUS B-mode images with two\nmeasurements, namely Jaccard Measure (JM) and Hausdorff Distances (HD). The\nevaluation result shows that IVUS-Net outperforms the state-of-the-art lumen\nand media segmentation methods by 4% to 20% in terms of HD distance. IVUS-Net\nperforms well on images in the test set that contain a significant amount of\nmajor artifacts such as bifurcations, shadows, and side branches that are not\ncommon in the training set. Furthermore, using a modern GPU, IVUS-Net segments\neach IVUS frame only in 0.15 seconds. The proposed work, to the best of our\nknowledge, is the first deep learning based method for segmentation of both the\nlumen and the media vessel walls in 20 MHz IVUS B-mode images that achieves the\nbest results without any manual intervention. Code is available at\nhttps://github.com/Kulbear/ivus-segmentation-icsm2018","url_abs":"http://arxiv.org/abs/1806.03583v2","url_pdf":"http://arxiv.org/pdf/1806.03583v2.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":"ivus-net-an-intravascular-ultrasound","repo_url":"https://github.com/Kulbear/ivus-segmentation-icsm2018","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"3d-reconstruction","task_name":"3D Reconstruction"},{"task_slug":null,"task_name":"GPU"},{"task_slug":"segmentation","task_name":"Segmentation"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"fcn","method_name":"FCN"},{"method_slug":"max-pooling","method_name":"Max Pooling"}],"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}