文献类型：期刊文献

英文题名：Automatic Organ and Pan-cancer Segmentation in Abdomen CT: the FLARE 2023 Challenge

作者：Ma, Jun[1]; Zhang, Yao[2]; Gu, Song[3]; Ge, Cheng[4]; Wang, Ershuai[5]; Zhou, Qin[6]; Huang, Ziyan[7]; Lyu, Pengju[8,9]; He, Jian[10]; Wang, Bo[11]

机构：[1] The Department of Laboratory Medicine and Pathobiology, University of Toronto, Peter Munk Cardiac Center, UHN AI Hub, University Health Network, Vector Institute, Toronto, Canada; [2] Lenovo Research, Beijing, China; [3] The Department of Image Reconstruction, Nanjing Anke Medical Technology Co., Ltd., Nanjing, China; [4] Ocean University of China, Qingdao, China; [5] Department of Research and Development, ShenZhen Yorktal DMIT Co. LTD., Shenzhen, China; [6] Institute of Medical Robotics, School of Biomedical Engineering, Shanghai Jiao Tong University, Department of Computer Science and Engineering, East China University of Science and Technology, Shanghai, China; [7] Shanghai Jiao Tong University, Shanghai AI Laboratory, Shanghai, China; [8] City University of Macau, China; [9] Hanglok-Tech Co., Ltd., Hengqin, China; [10] The Department of Nuclear Medicine, Nanjing Drum Tower Hospital, Nanjing, China; [11] The Peter Munk Cardiac Center, University Health Network, Department of Laboratory Medicine and Pathobiology, Department of Computer Science, University of Toronto, Vector Institute, UHN AI Hub, University Health Network, Toronto, Canada

年份：2024

外文期刊名：arXiv

收录：EI(收录号：20240375129)

语种：英文

外文关键词：Benchmarking - Computerized tomography - Deep learning - Diagnosis - Diseases - Oncology

摘要：Organ and cancer segmentation in abdomen Computed Tomography (CT) scans is the prerequisite for precise cancer diagnosis and treatment. Most existing benchmarks and algorithms are tailored to specific cancer types, limiting their ability to provide comprehensive cancer analysis. This work presents the first international competition on abdominal organ and pan-cancer segmentation by providing a large-scale and diverse dataset, including 4650 CT scans with various cancer types from over 40 medical centers. The winning team established a new state-of-the-art with a deep learning-based cascaded framework, achieving average Dice Similarity Coefficient (DSC) scores of 92.3% for organs and 64.9% for lesions on the hidden multi-national testing set. The dataset and code of top teams are publicly available, offering a benchmark platform to drive further innovations https://codalab.lisn.upsaclay.fr/competitions/12239. ? 2024, CC BY-NC-ND.