{"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/a-new-ensemble-learning-framework-for-3d","title":"A New Ensemble Learning Framework for 3D Biomedical Image Segmentation","arxiv_id":"1812.03945","date":"2018-12-10","proceeding":null,"authors":["Hao Zheng","Yizhe Zhang","Lin Yang","Peixian Liang","Zhuo Zhao","Chaoli Wang","Danny Z. Chen"],"abstract":"3D image segmentation plays an important role in biomedical image analysis.\nMany 2D and 3D deep learning models have achieved state-of-the-art segmentation\nperformance on 3D biomedical image datasets. Yet, 2D and 3D models have their\nown strengths and weaknesses, and by unifying them together, one may be able to\nachieve more accurate results. In this paper, we propose a new ensemble\nlearning framework for 3D biomedical image segmentation that combines the\nmerits of 2D and 3D models. First, we develop a fully convolutional network\nbased meta-learner to learn how to improve the results from 2D and 3D models\n(base-learners). Then, to minimize over-fitting for our sophisticated\nmeta-learner, we devise a new training method that uses the results of the\nbase-learners as multiple versions of \"ground truths\". Furthermore, since our\nnew meta-learner training scheme does not depend on manual annotation, it can\nutilize abundant unlabeled 3D image data to further improve the model.\nExtensive experiments on two public datasets (the HVSMR 2016 Challenge dataset\nand the mouse piriform cortex dataset) show that our approach is effective\nunder fully-supervised, semi-supervised, and transductive settings, and attains\nsuperior performance over state-of-the-art image segmentation methods.","url_abs":"http://arxiv.org/abs/1812.03945v1","url_pdf":"http://arxiv.org/pdf/1812.03945v1.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":"a-new-ensemble-learning-framework-for-3d","repo_url":"https://github.com/HaoZheng94/Ensemble","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"3d-medical-imaging-segmentation","task_name":"3D Medical Imaging Segmentation"},{"task_slug":"ensemble-learning","task_name":"Ensemble Learning"},{"task_slug":"image-segmentation","task_name":"Image Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1812.03945","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}