Abstract
An inter-patient stratified train/val/test split (70/15/15) of 3,832 fetal head ultrasound images with YOLO-format polygon segmentation labels for three anatomical structures: Brain, Cavum Septi Pellucidi (CSP) and Lateral Ventricles (LV). The images come from the HC18 Grand Challenge (van den Heuvel et al., 2018) and the fetal head biometry annotation dataset of Alzubaidi et al. (2023). Every frame of a patient sits in exactly one partition, preventing the leakage that inflates scores when frames from one patient appear in both training and test sets; the split is seeded (seed = 42) for reproducibility.
Cite as
M. Fazri Nizar, “Inter-Patient Split Fetal Head Ultrasound Segmentation Dataset in YOLO Polygon Format,” Harvard Dataverse, version 1 (CC0), 2026.
@misc{nizar2026fetaldata,
author = {Nizar, M. Fazri},
title = {{Inter-Patient Split Fetal Head Ultrasound Segmentation Dataset in YOLO Polygon Format}},
publisher = {Harvard Dataverse},
year = {2026},
version = {V1},
doi = {10.7910/DVN/FKXHLL},
url = {https://doi.org/10.7910/DVN/FKXHLL}
}