A new artificial intelligence tool approved in the United States could significantly improve how doctors detect fetal abnormalities during pregnancy, helping address a long-standing gap in prenatal care.
A Smarter Way to Read Ultrasounds
The US Food and Drug Administration has cleared an AI-powered ultrasound assistant developed by American start-up BioticsAI for clinical use. The software works alongside existing ultrasound machines, analysing images in real time while scans are being performed.
Prenatal ultrasounds are a routine part of pregnancy care, offering doctors a window into how a baby is developing. However, many abnormalities affecting organs or limbs can still be missed. BioticsAI’s technology aims to support clinicians by highlighting potential issues as the scan happens, rather than after the fact.
How the Technology Supports Doctors
As an ultrasound image is captured, the software immediately assesses its quality and recommends adjustments if the view is unclear. It then checks whether all key fetal structures are visible and flags any missing or incomplete images.
Drawing on patterns learned from large global datasets, the AI can identify signs of potential problems, such as heart defects or limb abnormalities. Once the scan is complete, it compiles its findings into a report that clinicians can review, potentially cutting down documentation time by around eight minutes per patient.
The FDA’s clearance confirms that the software meets required safety and performance standards and can be used as a regulated medical device within existing healthcare systems.
Why Better Detection Matters
Congenital anomalies remain a significant challenge. According to European Commission data, nearly 24 in every 10,000 births across Europe involve major birth defects. Research from Cochrane, which reviewed studies covering more than seven million pregnancies, shows just how limited traditional screening can be.
An early ultrasound scan between 11 and 14 weeks detects only 38 percent of abnormalities. A later scan at 18 to 24 weeks improves detection to 51 percent. Using both scans together raises that figure to 84 percent, leaving room for improvement.
AI-assisted ultrasounds are gaining traction globally. In Europe, French companies such as Diagnoly and Sonio Detect have already received approval for similar tools that automatically identify fetal structures and heart conditions. Together, these technologies point to a future where prenatal screening is faster, more accurate and more supportive for both doctors and expectant parents.
