{"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/liver-segmentation-from-multimodal-images","title":"Liver Segmentation from Multimodal Images using HED-Mask R-CNN","arxiv_id":"1910.10504","date":"2019-10-23","proceeding":null,"authors":[],"abstract":"Precise segmentation of the liver is critical for computer-aided diagnosis\nsuch as pre-evaluation of the liver for living donor-based transplantation\nsurgery. This task is challenging due to the weak boundaries of organs,\ncountless anatomical variations, and the complexity of the background. Computed\ntomography (CT) scanning and magnetic resonance imaging (MRI) images have\ndifferent parameters and settings. Thus, images acquired from different\nmodalities differ from one another making liver segmentation challenging task.\nWe propose an efficient liver segmentation with the combination of\nholistically-nested edge detection (HED) and the Mask-region-convolutional\nneural network (R-CNN) to address these challenges. The proposed HED-Mask R-CNN\napproach is based on effective identification of edge maps from multimodal\nimages. The proposed system firstly applies a preprocessing step of image\nenhancement to get the 'primal sketches' of the abdomen. Then the HED network\nis applied to enhanced CT and MRI modality images to get a better edge map.\nFinally, the Mask R-CNN is used to segment the liver from edge map images. We\nused a dataset of 20 CT patients and 9 MR patients from the CHAOS challenge.\nThe system is trained on CT and MRI images separately and then converted to 2D\nslices. We significantly improved the segmentation accuracy of CT and MRI\nimages on a database with a Dice value of 0.94 for CT, 0.89 for T2-weighted MRI\nand 0.91 for T1-weighted MRI.","url_abs":"http://arxiv.org/abs/1910.10504v1","url_pdf":"http://arxiv.org/pdf/1910.10504v1.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":[],"tasks":[{"task_slug":"computed-tomography-ct","task_name":"Computed Tomography (CT)"},{"task_slug":"edge-detection","task_name":"Edge Detection"},{"task_slug":"image-enhancement","task_name":"Image Enhancement"},{"task_slug":"liver-segmentation","task_name":"Liver Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/segmentation-on","task":"Segmentation","dataset":"!(()&&!|*|*|","model":"HNN","rank_in_archive_order":1,"of":1,"metrics":{"10%":"20"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}