Researchers at the University of Ottawa's Faculty of Medicine are developing the use of a new artificial intelligence (AI)-based deep learning model as an assistive aid for the speedy and accurate reading of ultrasound images in a new proof-of-concept study led by Dr. Mark Walker.
The study's purpose was to show that deep-learning technology can facilitate early and accurate diagnosis of cystic hygroma from first trimester ultrasound imaging. Cystic hygroma is an embryonic disease that causes an aberrant development of the lymphatic vascular system. It is an uncommon and possibly fatal condition characterized by fluid accumulation around the skull and neck. There are some treatments available that have shown reasonable success rates, but like so many other defects and diseases, early detection can make all the difference.
Dr. Walker, co-founder of The Ottawa Hospital's OMNI Research Group (Obstetrics, Maternal and Newborn Investigations), and his research team wanted to see how well AI-driven pattern detection might detect the abnormality during an imaging session.
Potential early detection for other birth defects
“What we demonstrated was in the field of ultrasound we’re able to use the same tools for image classification and identification with a high sensitivity and specificity,” said Dr. Walker, who believes their approach might be applied to other fetal anomalies generally identified by ultrasonography.
The results were published in PLOS ONE, an open access peer-reviewed journal. The full details are available from the Faculty of Medicine.
