Abstract
The 2024 MICCAI COSAS International Challenge, jointly organized by the School of AI and Advanced Computing at Xi'an Jiaotong-Liverpool University (XJTLU), Ruijin Hospital Affiliated to Shanghai Jiao Tong University, National Engineering Laboratory for Big Data System Computing Technology at Shenzhen University, Guangxi Medical University, and Shanghai HISTO Pathology Diagnostic Center, successfully concluded on October 5th during the MICCAI Conference on Medical Image Computing and Computer-Assisted Intervention in Marrakesh, Morocco. Dr. Jingxin Liu, the challenge chair, gave a summary of the event and presented awards to the participants.
由西交利物浦大学人工智能与先进计算学院联合上海交通大学附属瑞金医院、深圳大学大数据系统计算技术国家工程实验室、广西医科大学以及上海市衡道病理医学诊断中心共同主办的2024年MICCAI COSAS国际挑战赛，于10月5日在摩洛哥马拉喀什举办的MICCAI国际医学图像计算与计算机辅助干预会议上圆满落幕。挑战赛主席刘净心博士主持了赛事总结并为获奖选手颁发奖项。
竞赛回顾
Competition review
This challenge, which spanned five months, attracted more than 250 teams and individuals worldwide. The AIMI team from the University of Erlangen-Nuremberg in Germany won the championship in both tasks. Teams from the Institute of Computing Technology at the Chinese Academy of Sciences, Shengmei Bio, and Amaranth Medical Analytics from the United States won the second and third places in both tasks, respectively.
此次挑战赛历时5个月，吸引了来自全球超过250支团队和个人参赛。最终，德国埃尔兰根-纽伦堡大学的AIMI团队在两项任务中获得了冠军，而中科院计算所、圣美生物以及美国Amaranth Medical Analytics团队分别获得两项任务的亚军与季军。
A total of 800 finely annotated images were used in the competition, and all the data will be made publicly available after the event, with the hope of further advancing research and development in the field of computational pathology. Students from XJTLU also gained valuable experience through the challenge. The SURF team, led by Dr. Jingxin Liu, compared the generalization capabilities of existing deep learning models on the challenge dataset during the early stages of the competition. The related results have already been submitted to the ICBIP 2024 conference. PhD student Biwen Meng continued her research on the first task, adenocarcinoma segmentation across different organs, with the latest findings submitted to ISBI 2025. Furthermore, a summary paper for the challenge is currently being prepared.
本次比赛使用了800张精细标注的数据，所有数据将在赛后全部开源，以期进一步推动计算病理学领域的研究与发展。通过此次竞赛，西交利物浦大学的学生也得到了宝贵的实践机会。由刘净心博士带领的SURF团队在挑战赛数据库上对比了现有的深度学习模型的泛化能力，其相关成果已整理成文并提交至ICBIP 2024会议。博士生孟碧雯针对首项任务——跨器官腺癌分割展开了深入研究，其最新成果已投稿至ISBI 2025。此外，关于本次竞赛的总结论文也正在撰写中。
竞赛总结
Summary
The successful organization of this challenge not only propelled global research in computational pathology but also provided a platform for teams to showcase their work and exchange ideas, offering valuable insights and support for future research in medical image computing. In addition, this challenge is also the first time that XJTLU Entrepreneur College (Taicang) has held the International Medical Imaging Challenge as an organizer, which fully demonstrates XJTLU’s distinctive education model of industry, teaching and research integration.
本次挑战赛的成功举办不仅推动了全球计算病理学的研究，也为参赛团队提供了展示与交流的平台，为未来的医学图像计算研究提供了重要的参考和支持。此外，此次挑战赛也是西浦创业家学院（太仓）第一次以主办方的身份举办的国际医学图像挑战赛，充分展现了西浦产业、教学和研究融合的特色教育模式。
人工智能与先进计算学院，由处于人工智能和计算发展前沿的龙头企业参与创建。因此，学院拥有与人工智能行业技术研发紧密相关的稳健商业发展战略。学院在创业精神的支持下，开展行业主导的研究，作为学院各项工作的核心。
XJTLU_AIAC
欢迎关注
西浦太仓人工智能与先进计算学院
了解更多学院相关信息
Scan QR code to know more about AIAC!
