A newly approved artificial intelligence system could change how doctors spot fetal abnormalities, addressing a long-standing problem in prenatal care where many issues go undetected.
Bringing AI Into the Scan Room
The United States Food and Drug Administration has cleared an AI-powered software developed by American start-up BioticsAI for use in clinical settings. The technology works alongside standard ultrasound machines, analysing images as they are captured and alerting clinicians to potential concerns during routine pregnancy scans.
Ultrasound examinations are a cornerstone of prenatal care, offering insights into a baby’s development before birth. While they can reveal a range of structural problems, many abnormalities affecting organs or limbs are still missed. The new software aims to act as an extra set of eyes, supporting doctors as they perform scans.
How the Software Works
As soon as an ultrasound image appears on screen, the AI evaluates its quality and suggests adjustments to improve clarity if needed. It then checks whether all key parts of the fetus are visible, flagging any missing views that may require further imaging.
By comparing the scan against patterns drawn from large international datasets, the system can highlight signs of potential issues such as heart defects or limb abnormalities. After the examination, it produces a summary report that doctors can review, a feature developers say can save around eight minutes per patient by reducing paperwork.
The FDA’s approval confirms the software meets medical device standards and can be safely integrated into existing healthcare systems.
Why Better Detection Is Needed
Birth defects remain a significant challenge worldwide. Data from the European Commission show that major congenital anomalies occur in nearly 24 out of every 10,000 births across Europe.
Research from Cochrane, which analysed studies involving more than seven million pregnancies, highlights the limits of current screening methods. An early ultrasound between 11 and 14 weeks detects just 38 percent of abnormalities, while a mid-pregnancy scan at 18 to 24 weeks finds about 51 percent. When both scans are used together, detection improves to 84 percent, but gaps remain.
Europe is also embracing AI-assisted screening. French companies Diagnoly and Sonio Detect have gained approval for similar tools that automatically identify fetal structures and heart conditions. Together, these advances suggest AI could play a growing role in making prenatal care more accurate and reassuring for families.
