Researchers at the University of Ottawa are pioneering the application of deep learning (DL) in prenatal care. DL’s pattern recognition capability is being leveraged to diagnose fetal anomalies in the first trimester.
Maternity care is complicated. Each year, about 900,000 women experience a range of physical, psychological, emotional and personal changes brought by pregnancy in England. Pregnancy is a major event in the lives of women, and one fraught with unique challenges.
One of the key challenges facing prenatal care is the occurrence of birth defects during early fetal development. According to the Centers for Disease Control and Prevention, most birth defects occur during the first trimester—a period wherein the organs of the baby begin to take form. Fortunately, it is possible to treat these defects in utero with timely diagnosis and medical intervention.
New research led by the University of Ottawa aims at mitigating the risk of the occurrence of birth defects in the first trimester. By leveraging deep learning (DL) algorithms for quick and precise assessment of ultrasound scans, researchers are looking to identify and diagnose cystic hygroma in fetuses—a potentially life-threatening birth defect.
“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,” commented Dr Mark C. Walker, the leader of the research initiative and the first author of the study.
Walker is optimistic that the deep learning-based approach can be utilized to detect other fetal anomalies when coupled with ultrasonographical methods.
Artificial intelligence (AI) technology is maturing and becoming more reliable.
The list of ways AI is assisting healthcare professionals is expanding. Each day researchers bring yet another innovative tool that enables faster diagnosis, improved patient outcomes, and better quality of life for people.