Neural network for detecting signs of spina bifida in the fetus during ultrasound examination of pregnant women

The main articles are:

• Neural networks (neural networks)
• Artificial Intelligence in Medicine
• Artificial intelligence in radiology

2024: Announcement of a neural network to detect signs of spina bifida in the fetus in an ultrasound study of pregnant women

On August 19, 2024, Yandex announced the development of a neural network that will help doctors detect the symptoms of spina bifida, a severe congenital disease of the central nervous system in children, during an ultrasound study of pregnant women. This congenital pathology is complex diagnose as it occurs once per thousand newborns. It often leads to severe disability. With the help of technology, medical specialists will be able to see signs of this disease at an earlier date and send the patient for additional examination. The solution is available free of charge to all doctors and medical experts on the website of the Spin Bifida Foundation, which initiated the first such project in Russia.

source = Yandex

According to the company, the technology can be used to develop other solutions for medicine. The medical specialist can upload the ultrasound results taken during the first screening to the solution. The solution will automatically analyze the picture and help the doctor determine the likelihood with which the signs of spina bifida are present in the image. If there is a possibility of pathology, the doctor will be able to promptly send the patient for additional examination to large perinatal centers in Russia in order to prescribe further treatment.

А что цитируем-то? comments Liliyana Chugunova, Candidate of Medical Sciences, Senior Researcher, Department of Ultrasound and Functional Diagnostics [[NMIC AGP V I Kulakova FSBI Ministry of Health of Russia - Scientific Center for Obstetrics, of Gynecology and Perinatology named after V.I. Kulakova|FSBI "NMIC AGP named after V.I. Kulakova" Ministry of Health of Russia]]

source = Yandex

For machine learning, a datacet of about 6 thousand impersonal images of ultrasound of pregnant women was used, collected by specialists from the National Medical Research Center (NMITs) of obstetrics, gynecology and perinatology named after academician V.I. Kulakov. Data markup and medical methodology of the project were led by ultrasound doctors from the Kulakov National Medical Research Center, experts in the field of anomaly detection fetal central nervous system. Yandex Cloud specialists, together with students from the Yandex School of Data Analysis, created a solution architecture using several types of neural networks to detect objects, classify them and determine the likelihood of pathology. Based on the dataset, SHAD students taught models and developed a web interface that is convenient for doctors.

Training took place on the cloud platform using the Yandex DataSphere service. Yandex Cloud provided computing power and help from architects for the project for free.

source = Yandex

{{quote 'author
= told Anna Lemyakina, director of national strategic projects Yandex Cloud'We reacted very responsibly to the initiative of the fund and hope that neural networks will serve as a good auxiliary tool for doctors. The experience gained can be used in others medical projects.}}

As a result of cooperation with Yandex and the Kulakov National Medical Research Center, we managed to implement the project at the junction of IT and healthcare. We see great potential in cooperation with machine learning specialists to implement such complex projects. told Inna Inyushkina, founder of the Spin Bifida Foundation

The project continues: the next stage is the collection and processing of data at the Kulakov National Medical Research Center for further training of the model and optimization of the quality of its work. Participants also posted the development code in OpenSource to connect active members of the IT community to the development. They will be able to use these technologies to create other services for supporting medical decision-making.