In Kazakhstan, AI is being tested for the early detection of congenital spinal abnormalities in fetuses.

Yandex Qazaqstan is launching a pilot project to detect signs of a rare fetal spinal pathology at an early stage of pregnancy. The project uses an AI system for analyzing ultrasound images, built on Yandex Cloud technologies. The Karaganda region has been selected as the pilot site.

Yandex Qazaqstan and the Karaganda Region Department of Healthcare announce the launch of a joint pilot project for the early detection of spina bifida, a congenital developmental disorder of the spine and spinal cord. The AI system for recognizing ultrasound images was created based on Yandex Cloud technologies and the expertise of the Center for Social Technologies. The initiative received support from the Akimat of the Karaganda Region, and on February 20, a memorandum of cooperation was signed between the region and Yandex Qazaqstan. The document was signed by the Deputy Akim of the Karaganda Region, Oraz Kaysaevich Taurbekov, and Arystanbek Joldasbekov, General Director of Yandex Cloud in Kazakhstan and Central Asia.

Spina bifida is a congenital condition in which the fetus's spine and spinal cord form incorrectly. In the most severe cases, this leads to paralysis and dysfunction of internal organs. The earlier the disease is detected, the higher the chances of a successful operation — it can be performed even in utero before the 26th week of pregnancy. Additionally, the pathology is rare, occurring in approximately one in a thousand newborns, so many doctors have not encountered it during their careers.

The idea for the project to apply AI to detect signs of the disease belongs to the "Spina Bifida" foundation, which systematically helps children and adults with spina bifida. It promotes early diagnosis of the disease during pregnancy, supports families and orphanages, enabling people with this diagnosis to live active and fulfilling lives. Experts from the V.I. Kulakov National Medical Research Center of Obstetrics, Gynecology and Perinatology play an important role in the project — one of the key scientific and clinical centers in the field of prenatal diagnosis, where one of the largest data sets has been accumulated. The technology has already been tested at the V.I. Kulakov National Medical Research Center of Obstetrics, Gynecology and Perinatology and in several other clinics in Moscow, St. Petersburg, and Yekaterinburg.

The use of computer vision improves the accuracy of detecting signs of pathology in ultrasound images at early stages. At the same time, the quality of data plays a key role in the successful training of neural networks. Cloud technologies from Yandex Cloud are used for storing and processing data, as well as for training neural networks. They allow working with any volume of data necessary for training and operating models, as well as handling high ML workloads.

Arystanbek Joldasbekov, General Director of Yandex Cloud in Kazakhstan and Central Asia
The Center for Social Technologies at Yandex Qazaqstan addresses issues relevant to all Kazakhstani citizens. These include environmental challenges, preservation of cultural heritage, and now projects in the field of medicine. The results we will obtain during the pilot project with the Karaganda Region Department of Healthcare are valuable not only in the context of this project but also for future initiatives involving the application of cloud technologies in medicine.

Anna Lemyakina, Head of the Center for Social Technologies at Yandex Qazaqstan
Yandex Cloud technologies are used in many socially significant projects, and early detection of spina bifida signs is another important task we can help with. The neural network will become a doctor's assistant, providing decision support. The doctor's expertise is necessary for the final assessment and subsequent decisions, while cloud and AI technologies allow recognizing the smallest details in images that the human eye might miss.

Oraz Taurbekov, Deputy Akim of the Karaganda Region, Kaysaevich
Since the technology has already been successfully tested, we hope that the tool will soon become operational for Kazakhstani doctors.

The pilot project is launching in the Karaganda Region, and its results may later serve as a basis for scaling the solution across all of Kazakhstan.

Center for Social Technologies at Yandex Qazaqstan was established in 2024 and is aimed at supporting socially significant projects in four areas: science and education, healthcare, ecology, and culture. The Center provides projects with expert assistance, access to Yandex Cloud's cloud technologies and machine learning tools, as well as resources for promoting solutions.