Butterfly Network Wins First FDA Clearance for a Blind-Sweep Ultrasound AI That Estimates Gestational Age Without a Trained Sonographer
The FDA has cleared Butterfly Network's Gestational Age Tool, the first blind-sweep ultrasound AI authorized for U.S. marketing, letting any healthcare worker estimate fetal age in under two minutes.
Overview
The U.S. Food and Drug Administration has cleared Butterfly Network’s fully automated Gestational Age (GA) Tool, making it the first blind-sweep ultrasound AI authorized for the U.S. market, according to a press release published on March 30, 2026. The tool is integrated into Butterfly’s handheld semiconductor-based ultrasound platform and enables any healthcare worker — not just trained sonographers — to estimate fetal gestational age in under two minutes.
The clearance marks a significant step toward closing obstetric imaging gaps in rural and underserved communities, both in the United States and globally. The GA Tool is already deployed in research settings across 12 countries, including clinical use in Malawi and Uganda, supported in part by a grant from the Bill & Melinda Gates Foundation.
How the Blind-Sweep Method Works
Conventional fetal ultrasound requires a trained sonographer to identify specific anatomical landmarks — head circumference, femur length, abdominal circumference — and manually measure them to calculate gestational age. This process, known as biometry, takes specialized training and equipment that remain scarce in low-resource settings.
The GA Tool eliminates both requirements through a three-step automated process, as described in the company’s announcement: the operator enters the patient’s fundal height, applies gel, and performs six guided sweeps across the abdomen. The operator does not need to identify fetal anatomy, interpret the images, or even look at the screen during the scan. A deep-learning model processes the video in the background and returns a gestational age estimate for patients between 16 and 37 weeks of pregnancy.
Dr. Jeffrey Stringer, Clarke-Pearson Distinguished Professor of Obstetrics and Gynecology at the University of North Carolina at Chapel Hill, developed the underlying AI models with his team. “Our goal has been to decouple image acquisition from interpretation, making high-quality obstetric assessment accessible far beyond traditional care settings,” Stringer said. UNC licensed the innovation to Butterfly Network, which integrated it into the Butterfly app for global distribution.
Clinical Validation
The AI behind the GA Tool was validated in a study published in NEJM Evidence in 2022, which enrolled 4,695 pregnant volunteers in North Carolina and Zambia between September 2018 and June 2021. The neural network achieved a mean absolute error of 3.9 days, compared to 4.7 days for standard sonographer-performed biometry — a statistically significant difference of 0.8 days favoring the AI model (P<0.001).
The results held across geographies. In North Carolina, the AI outperformed biometry by 0.6 days; in Zambia, the gap widened to 1.0 day. In a subset of pregnancies conceived through in vitro fertilization, where the true gestational age is known precisely, the AI achieved 2.8 days of mean absolute error versus 3.6 days for biometry. When untrained operators used low-cost handheld devices, the AI still produced estimates within 4.9 days — close to the 5.4-day error seen with trained sonographers using standard equipment.
The commercial GA Tool has since been trained on more than 21 million images spanning diverse patient demographics and clinical environments, according to Butterfly Network.
Global and Domestic Expansion
Accurate gestational age estimation is foundational to prenatal care — it determines when to screen for fetal anomalies, when to induce labor, and how to manage preterm birth. The World Health Organization estimates that 92 percent of maternal and neonatal deaths occur in low- and middle-income countries, where access to trained sonographers and imaging equipment is limited.
The GA Tool is already in use in Malawi and Uganda as part of maternal health initiatives funded by the Gates Foundation. Steve Cashman, Butterfly Network’s Chief Business Officer, outlined three primary use cases for the cleared device: emergency departments where obstetricians are unavailable, high-mortality countries with sonographer shortages, and rural U.S. communities with limited access to prenatal imaging, according to the press release.
Sachita Shah, Butterfly Network’s Vice President of Global Health, said the tool was designed to “help transform maternal health” by reaching populations that conventional ultrasound infrastructure has failed to serve.
The FDA clearance paves the way for Butterfly to expand distribution across additional sub-Saharan African regions and to begin targeting the estimated 2.2 million women in the United States who live in counties classified as maternity care deserts — areas with no obstetric providers, no birth facilities, and no certified nurse-midwives.