Yandex, a Russian company, has developed an innovative IT solution utilizing machine learning. This technology enables the detection of spina bifida symptoms within the fetus during ultrasound examinations in expectant mothers. This is a severe congenital disease of the central nervous system that occurs in one in a thousand newborns. The solution is available free of charge to all doctors and medical experts on the website of the Spina Bifida Foundation, which has launched this project in Russia.
With the help of technology, medical specialists will be able to detect signs of spina bifida even before the baby is born and promptly refer the patient for further examination. To do this, the doctor must provide the neural network with the ultrasound results taken during the first screening. The program will automatically analyze the image and determine the probability of spina bifida. A unique dataset of approximately 6,000 anonymized ultrasound images of pregnant women, collected by specialists from the National Medical Research Center for Obstetrics, Gynecology and Perinatology named after Academician V.I.Kulakov of the Ministry of Healthcare, was utilized for machine learning.
