文献类型：期刊文献

英文题名：IUGC: A benchmark of landmark detection in end-to-end intrapartum ultrasound biometry

作者：Bai, Jieyun[1,2,3];Tang, Yitong[3];Liu, Xiao[4];Hu, Jiale[4];Li, Yunda[4];Chen, Xufan[4];Wang, Yufeng[4];Ma, Chen[5];Li, Yunshu[5];Guo, Bowen[5];Jiao, Jing[5];Huang, Yi[5];Wang, Kun[6];Li, Lifei[6];Ma, Yuzhang[6];Han, Xiaoxin[6];Shao, Haochen[6];Yang, Zi[7];Liu, Qingchen[7];Hu, Yuchen[7];Kuang, Jingfan[7];Song, Shanglin[7];Krishna, Anirvan[8];Khan, Zaid Ahmed[8];Li, Zelan[9];Zhang, Zhengyang[9];Zhang, Hansen[9];Cheng, Yan[9];Zhang, Xuezhi[10];Chen, Xi[10];Yan, Hao[10];Tong, Lyuyang[10];Du, Bo[10];Deng, Bo[3];Chen, Yu[3];Peng, Zilun[3];Rezaei, Saeid[11];Gan, Jie[12];Cai, Weidong[12];Wang, Fangyijie[13];Curran, Kathleen M.[13];Silvestre, Guenole[13];Khobo, Isaac[14];Lu, Yaosheng[3,9,10];Ni, Dong[15];Huang, Yuxin[16];Yaqub, Mohammad[17];Ma, Jun[9,10,18];Lekadir, Karim[19];Li, Shuo[20]

第一作者：Bai, Jieyun

通信作者：Bai, JY[1];Bai, JY[2];Bai, JY[3];Huang, YX[4];Li, S[5]

机构：[1]Jinan Univ, Affiliated Hosp 1, Dept Cardiovasc Surg, Guangzhou, Peoples R China;[2]Univ Auckland, Auckland Bioengn Inst, Auckland, New Zealand;[3]Jinan Univ, Sch Informat Sci & Technol, Guangzhou, Peoples R China;[4]Nanyang Inst Technol, Nanyang, Peoples R China;[5]Fudan Univ, Shanghai, Peoples R China;[6]Gansu Univ Chinese Med, Lanzhou, Peoples R China;[7]Lanzhou Univ First Hosp, Lanzhou, Peoples R China;[8]Indian Inst Technol Kharagpur, Kharagpur, India;[9]Northeastern Univ, Shenyang, Peoples R China;[10]Wuhan Univ, Sch Comp Sci, Wuhan, Peoples R China;[11]Univ Coll Cork, Cork, Ireland;[12]Univ Sydney, Sch Comp Sci, Sydney, Australia;[13]Univ Coll Dublin, Sch Med, Dublin, Ireland;[14]Univ Cape Town, Cape Town, South Africa;[15]Shenzhen Univ, Shenzhen, Peoples R China;[16]Southern Med Univ, Zhujiang Hosp, Obstet & Gynecol Ctr, Guangzhou, Peoples R China;[17]Mohamed Bin Zayed Univ Artificial Intelligence, Dept Machine Learning, Abu Dhabi, U Arab Emirates;[18]Univ Hlth Network, Toronto, ON, Canada;[19]Artificial Intelligence Med Lab BCN AIM, Barcelona, Spain;[20]Case Western Reserve Univ, Biomed Engn, Cleveland, OH 44106 USA

第一机构：Jinan Univ, Affiliated Hosp 1, Dept Cardiovasc Surg, Guangzhou, Peoples R China

通信机构：[1]corresponding author), Jinan Univ, Affiliated Hosp 1, Dept Cardiovasc Surg, Guangzhou, Peoples R China;[2]corresponding author), Univ Auckland, Auckland Bioengn Inst, Auckland, New Zealand;[3]corresponding author), Jinan Univ, Sch Informat Sci & Technol, Guangzhou, Peoples R China;[4]corresponding author), Southern Med Univ, Zhujiang Hosp, Obstet & Gynecol Ctr, Guangzhou, Peoples R China;[5]corresponding author), Case Western Reserve Univ, Biomed Engn, Cleveland, OH 44106 USA.

年份：2026

卷号：110

外文期刊名：MEDICAL IMAGE ANALYSIS

收录：;EI(收录号：20260519983550);WOS:【SCI-EXPANDED(收录号:WOS:001679915000001)】；

基金：This study was supported by project grants from the Guangzhou Science and Technology Planning Project (2025B03J0127) , the Natural Science Foundation of Guangdong Province (2024A1515011886) , the National Institute of Hospital Administration (No. YLXX24AIA006) , Sichuan Provincial Cross-Regional Innovation Cooperation Project (No. 2025YFHZ0326) , and Key Research and Development Program of Guangxi Province (No. 2023AB22074 and 2024AB04027) , the National Natural Science Foundation of China (61901192) , the Guangzhou Municipal Science and Technology Bureau Guangzhou Key Research and Development Program (2024B03J1283 and 2024B03J1289) , the High-end Foreign Experts Recruitment Plan of China (H20240205) , the China Scholarship Council (202206785002) , and the European Research Council (ERC) under the Horizon Europe programme (AIMIX project-Grant Agreement No. 101044779) .

语种：英文

外文关键词：Fetal ultrasound; Fetal biometry; Intrapartum ultrasound; Artificial intelligence; Angle of progression; Isuog; Perinatal medicine; Intrapartum care; Landmark detection; Foundation models; Semi-supervised learning; Challenge

摘要：Accurate intrapartum biometry plays a crucial role in monitoring labor progression and preventing complications. However, its clinical application is limited by challenges such as the difficulty in identifying anatomical landmarks and the variability introduced by operator dependency. To overcome these challenges, the Intrapartum Ultrasound Grand Challenge (IUGC) 2025, in collaboration with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), was organized to accelerate the development of automatic measurement techniques for intrapartum ultrasound analysis. The challenge featured a large-scale, multi-center dataset comprising over 32,000 images from 24 hospitals and research institutes. These images were annotated with key anatomical landmarks of the pubic symphysis (PS) and fetal head (FH), along with the corresponding biometric parameter-the angle of progression (AoP). Ten participating teams proposed a variety of end-to-end and semi-supervised frameworks, incorporating advanced strategies such as foundation model distillation, pseudo-label refinement, anatomical segmentation guidance, and ensemble learning. A comprehensive evaluation revealed that the winning team achieved superior accuracy, with a Mean Radial Error (MRE) of 6.53 +/- 4.38 pixels for the right PS landmark, 8.60 +/- 5.06 pixels for the left PS landmark, 19.90 +/- 17.55 pixels for the FH tangent landmark, and an absolute AoP difference of 3.81 +/- 3.12 degrees This top-performing method demonstrated accuracy comparable to expert sonographers, emphasizing the clinical potential of automated intrapartum ultrasound analysis. However, challenges remain, such as the trade-off between accuracy and computational efficiency, the lack of segmentation labels and video data, and the need for extensive multi-center clinical validation. IUGC 2025 thus sets the first benchmark for landmark-based intrapartum biometry estimation and provides an open platform for developing and evaluating real-time, intelligent ultrasound analysis solutions for labor management.