The growth and development of the fetal head and brain during pregnancy are crucial for the child’s long-term health and well-being. However, fetal ultrasound images are prone to variations due to patient-specific factors and image issues such as signal dropouts, artifacts, and missing boundaries. Hence, there is a need for automatic methods to ensure accurate and consistent measurements of fetal head and brain growth
Hyperparameter tuning performed using KerasTuner
Paper: Automating UNet Architecture Search for Fetal Head Segmentation
Accepted at TENCON 2026; author’s accepted version
Application: https://aaylmao-hc-prediction.hf.space/