AI breakthrough reveals pregnancy risks doctors miss – and it could save thousands of lives
Every year, millions of pregnant women undergo routine ultrasounds to track their baby’s growth.
But what if those scans are missing vital clues? That’s exactly what a group of Danish researchers has discovered.
AI sees what humans can’t
Enter Prenaital – a spinout company formed by researchers at the Technical University of Denmark (DTU) and the University of Copenhagen.
Their AI model, still in development, has shown it can detect up to 35 % more high-risk pregnancies than traditional ultrasound interpretation alone.
Trained on more than 10,000 ultrasound images from Danish hospitals, the AI uses deep neural networks to uncover patterns in fetal development.
These include brain structures, fat distribution, and tissue composition – subtle signs that trained clinicians might miss.
Instead of just measuring head or abdominal circumference, the model analyzes the entire image, unlocking insights previously buried in pixels.
The AI then generates a diagnostic report to assist clinicians in making earlier and more accurate decisions.
Coming to clinics soon
The technology is not yet available for clinical use, but the first product – an AI tool for fetal growth scans – is expected to launch in 2026 in Denmark.
The team behind the technology includes leading medical professionals and AI experts, such as Professor Martin G. Tolsgaard from Rigshospitalet.
He’s spent years researching how AI support affects diagnostic safety – and he’s convinced it will make a difference.
This article is based on information from Via Ritzau.