Ultrasound AI has received De Novo clearance from the US Food and Drug Administration for its flagship Delivery Date AI technology, a cloud-based Software as a Medical Device (SaMD) designed to determine a Predicted Delivery Date (PDD) using standard ultrasound images.

The technology integrates seamlessly into existing obstetrics (OB) and maternal-fetal medicine (MFM) prenatal workflows and provides real-time delivery date predictions to assist clinical teams in decision-making. Delivery Date AI is intended as an adjunctive tool in pregnancies where traditional dating methods—such as last menstrual period or ultrasound-based gestational age estimates—are unreliable.

Trained on millions of de-identified ultrasound images across diverse patient populations and clinical settings, the system employs an ensemble of deep-learning neural networks to analyze full ultrasound images, including fetal and maternal characteristics associated with delivery timing. By converting imaging data into actionable insights, the platform aims to support individualized care planning and follow-up.

The technology was evaluated in the PAIR (Perinatal Artificial Intelligence in Ultrasound) Study, published in The Journal of Maternal-Fetal & Neonatal Medicine, in collaboration with the University of Kentucky. The study, involving more than 5,700 patients, reported a 0.92 R² value in predicting days to delivery using standard ultrasound images, underscoring the model’s accuracy across varied clinical settings.

Robert Bunn, President and Founder of Ultrasound AI, described the clearance as a milestone in advancing earlier and more informed decision-making aimed at reducing the burden of preterm birth. Drza Nathan Fox, a board-certified OBGYN and maternal-fetal medicine specialist, noted that improved insight into pregnancy progression could help clinicians better determine how and when to intervene.

Delivery Date AI is compatible with most existing ultrasound systems, can be deployed within minutes, and generates results within seconds of image upload. The company said the scalable solution is designed for use across high-volume hospitals, resource-constrained clinics, and underserved obstetric regions, with potential to reduce costs associated with preterm birth.