What question did this study set out to answer?

Develop an open-access dataset to enable AI-assisted assessment of fetal growth through precise biometric measurements.

April 16, 2026Scientific Reports0 citationsOpen Access

Multicentre Benchmark Dataset for Fetal Ultrasound Biometry

A multicentre benchmark dataset for comprehensive landmark-based fetal ultrasound biometry

View Full Paper

Ask AI

Bookmark

Authors

Benchmark dataset provides expert-annotated fetal ultrasound images, aiding AI-assisted growth assessments in diverse settings.

Develop an open-access dataset to enable AI-assisted assessment of fetal growth through precise biometric measurements.
Compiled a dataset of 4,513 de-identified ultrasound images from 1,904 subjects across four clinical sites.
Annotated anatomical landmarks critical for fetal biometric measurements in standard planes.
Provided code and evaluation metrics for reproducibility and comparison across different methods.
Dataset includes various biometric measurements like head diameters, abdominal dimensions, and femoral length.
Training on single-site data leads to performance overestimation compared to multicentre evaluation.
First publicly available multi-device, landmark-annotated dataset for fetal biometry, aiding generalization studies.

Vece et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

April 16, 2026Scientific Reports0 citationsOpen Access

A multicentre benchmark dataset for comprehensive landmark-based fetal ultrasound biometry

View Full Paper

Ask AI

Bookmark

Authors

Benchmark dataset provides expert-annotated fetal ultrasound images, aiding AI-assisted growth assessments in diverse settings.

Develop an open-access dataset to enable AI-assisted assessment of fetal growth through precise biometric measurements.
Compiled a dataset of 4,513 de-identified ultrasound images from 1,904 subjects across four clinical sites.
Annotated anatomical landmarks critical for fetal biometric measurements in standard planes.
Provided code and evaluation metrics for reproducibility and comparison across different methods.
Dataset includes various biometric measurements like head diameters, abdominal dimensions, and femoral length.
Training on single-site data leads to performance overestimation compared to multicentre evaluation.
First publicly available multi-device, landmark-annotated dataset for fetal biometry, aiding generalization studies.

Vece et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark