A Foundational EDM2-Based Generative Model for High-Resolution Synthetic Fetal Ultrasound Imaging from Open Datasets

arXiv:2608.05471 · eess.IV, cs.CV, cs.LG, physics.med-ph · Submitted 2026-08-05 · Read on arXiv

Harvey Mannering, Yilin Zhang, Ziao Liu, Zhiwu Huang, Jacqueline Matthew, Miguel Xochicale

eess.IV, cs.CV, cs.LG, physics.med-ph

Submitted: 2026-08-05

Comments: Short paper for the 30th Conference on Medical Image Understanding and Analysis

Code: https://github.com/xfetus/fetal-ultrasound-edm2

Project page: https://xfetus.github.io/fetal-ultrasound-edm2-survey-2026

License: http://creativecommons.org/licenses/by-sa/4.0/

The gist: Prenatal ultrasound imaging is key for assessing fetal health, but AI progress is limited by scarce, privacy-restricted, and hard-to-annotate datasets.

Terminology

Abstract

Prenatal ultrasound imaging is key for assessing fetal health, but AI progress is limited by scarce, privacy-restricted, and hard-to-annotate datasets. We propose a high-resolution fetal ultrasound synthesis framework based on the EDM2 diffusion architecture, trained on multiple public datasets to generate 512x512 images across six anatomical classes. Our method achieved improved image quality with lower FID scores and enhanced downstream fetal plane classification, reaching 93.36% ensemble accuracy after fine-tuning, surpassing real-data-only training. Clinical evaluation by an experienced fetal ultrasound specialist (10+ years) on 100 images yielded a mean realism score of 2.67/5, with real images rated higher than synthetic. Artefacts included smoothing, speckle irregularities, and anatomical inconsistencies. Code, data, models and other resources to reproduce this work are available at https://github.com/xfetus/fetal-ultrasound-edm2.

Sources

Related papers