Despite considerable progress in maternal healthcare, maternal and perinatal deaths remain high in low-to-middle income countries. Fetal ultrasound is an important component of antenatal care, but shortage of adequately trained healthcare workers has limited its adoption. We developed and validated an artificial intelligence (AI) system that uses novice-acquired "blind sweep" ultrasound videos to estimate gestational age (GA) and fetal malpresentation. We further addressed obstacles that may be encountered in low-resourced settings. Using a simplified sweep protocol with real-time AI feedback on sweep quality, we have demonstrated the generalization of model performance to minimally trained novice ultrasound operators using low cost ultrasound devices with on-device AI integration. The GA model was non-inferior to standard fetal biometry estimates with as few as two sweeps, and the fetal malpresentation model had high AUC-ROCs across operators and devices. Our AI models have the potential to assist in upleveling the capabilities of lightly trained ultrasound operators in low resource settings.
@article{arxiv.2203.10139,
title = {AI system for fetal ultrasound in low-resource settings},
author = {Ryan G. Gomes and Bellington Vwalika and Chace Lee and Angelica Willis and Marcin Sieniek and Joan T. Price and Christina Chen and Margaret P. Kasaro and James A. Taylor and Elizabeth M. Stringer and Scott Mayer McKinney and Ntazana Sindano and George E. Dahl and William Goodnight and Justin Gilmer and Benjamin H. Chi and Charles Lau and Terry Spitz and T Saensuksopa and Kris Liu and Jonny Wong and Rory Pilgrim and Akib Uddin and Greg Corrado and Lily Peng and Katherine Chou and Daniel Tse and Jeffrey S. A. Stringer and Shravya Shetty},
journal= {arXiv preprint arXiv:2203.10139},
year = {2022}
}