Jieyun Bai's Challenge Paper "IUGC: A Benchmark of Landmark Detection in End-to-End Intrapartum Ultrasound Biometry"
Published in Bioengineering & Biotechnology
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°. 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.
I am an Associate Professor at Jinan University, a Joint Appointed Professor at the First Affiliated Hospital of Jinan University, and the Chief Scientist for Medical Artificial Intelligence at Guangzhou Lianyin Medical Technology Co., Ltd.
I was a postdoctoral researcher at the Auckland Bioengineering Institute, University of Auckland, working with Prof. Jichao Zhao from 2018 to 2019. I received my Ph.D. in Computer Science and Technology from Harbin Institute of Technology in 2017, under the supervision of Prof. Kuanquan Wang and Prof. Henggui Zhang, and I was also a visiting scholar at the Auckland Bioengineering Institute for one year.
My research focuses on AI-driven medical digital twins, ultrasound-based maternal–fetal monitoring, multimodal biomedical signal processing, and clinical decision support systems. I have published extensively on deep learning–based medical image and signal analysis in leading international conferences, including MICCAI, ISBI, EMBC, BIBM, and CinC, as well as in top-tier journals such as Medical Image Analysis (MIA), IEEE Transactions on Medical Imaging (TMI), IEEE Journal of Biomedical and Health Informatics (JBHI), and PLOS Computational Biology.
My team is dedicated to translational and data-driven medical AI research, leveraging large-scale clinical data collected from multi-center hospital systems. We actively develop open datasets, benchmark challenges, and open-source platforms to support reproducible and clinically relevant AI research.
I am a leading organizer in the international ultrasound AI community and currently serve as an Area Chair for MICCAI and ISBI Challenges. Through these initiatives, I have coordinated global collaborations among academia, hospitals, and industry, and have released large-scale, clinically curated datasets and standardized evaluation platforms to the community.
We are strongly committed to open science and have released more than 10 public datasets and code repositories on platforms such as Kaggle and Zenodo.
My group is currently recruiting postdoctoral fellows, research scientists, and graduate students in medical image analysis, medical digital twins, and AI-enabled maternal and cardiovascular healthcare.
Researchers and clinicians interested in medical AI and translational biomedical engineering are warmly welcome to contact me.
Please sign in or register for FREE
If you are a registered user on Research Communities by Springer Nature, please sign in