02/07/2025 Fangyijie Wang , Low-Cost Deep Learning for Fetal Growth Monitoring in Low-Resource Settings, Univ. College Dublin

16:00h (room 55.309)

Abstract:

Accurate fetal growth monitoring is essential for reducing maternal and neonatal complications, yet access to expert ultrasound interpretation remains limited in many low-resource settings. This talk presents our recent efforts to develop low-cost deep learning methods for fetal ultrasound analysis,
focusing on segmentation and gestational age (GA) estimation tasks. We explore lightweight architectures, semi-supervised learning techniques, and domain adaptation strategies that improve model performance under data scarcity and variability. The presentation will highlight contributions to
international challenges, and collaborations under the AIMIX project aimed at deploying AI tools in underserved clinical environments. Our goal is to build robust and efficient AI systems that support equitable maternal–fetal healthcare.

Fangyijie Wang is a PhD researcher at University College Dublin (ML-Labs), supervised by Dr. Kathleen Curran and Dr. Guénolé Silvestre, working on deep
learning for fetal ultrasound image analysis. He holds Master’s degrees in Statistics and Computer Science from UCD, and Bachelor's degrees from
Waterford Institute of Technology and Nanjing University of Information Science and Technology. Prior to his PhD, he worked as a Senior Data Engineer at
Verizon Connect and Citigroup. His research focuses on developing inclusive and efficient AI tools for maternal-fetal health in low-resource settings, with
publications in Medical Image Analysis, IEEE Access, MICCAI, and CVPR workshops. He has has earned awards in several challenges and hackathons.