Researchers at the University of Guelph have developed a new method using artificial intelligence (AI) to analyze ultrasound images, which could improve how doctors diagnose brain issues in newborns and premature babies.
The AI method could make it faster and easier to detect serious brain problems. Right now, about 20-30% of low-weight babies can develop bleeding in a part of the brain called the ventricles. If not diagnosed, this can lead to lifelong brain damage.
Currently, doctors use 2D ultrasound, which often produces blurry images, or 3D ultrasound, which is clearer but harder to analyze manually. The researchers found that using an AI method called semi-supervised learning (SSL) can help doctors see the shape and size of the brain’s ventricles more clearly, allowing for quicker and more accurate diagnoses.
Led by Dr. Eranga Ukwatta and PhD student Zachary Szentimrey, the team believes this AI method could help doctors start treatments sooner, leading to better outcomes for these vulnerable patients.
Source: educationnewscanada
