N. N. Petrov National Medical Research Center of Oncology has implemented a system based on YandexGPT models to automate the processing of clinical trial documents. Now, the initial data verification takes only a few minutes instead of weeks.
The system analyzes hundreds of pages of documents, finds errors, and suggests corrections. This helps doctors and pharmaceutical companies prepare accurate documents faster and coordinate research. Patients gain access to new medicines, medical devices, and treatments.
The development is planned to be used in oncology centers in BRICS countries and institutions conducting clinical trials.
Specialists from the local ethics committee (LEC) at the National Medical Research Center are already using the solution created by the Center for Technologies for Society Yandex Cloud (part of Yandex B2B Tech), Raft, and N. N. Petrov National Medical Research Center of Oncology.
The LEC is responsible for assessing the ethical aspects of each clinical trial. Its tasks include analyzing the risk-benefit ratio and ensuring compliance with the rights of study participants. The LEC considers about a thousand applications annually.
The development is a web service on the Yandex Cloud platform, where document scans are uploaded. YandexGPT 5 Pro technology analyzes documents, categorizes them into 10 types, extracts information, and performs automatic verification according to criteria developed by experts from N. N. Petrov National Medical Research Center of Oncology.
LEC experts track the status of applications in Yandex Tracker. In 95% of cases, the initial verification is performed without human intervention. Then, a specialist checks the accuracy of the data and makes a decision: approve or reject the application.
The neural network does not replace the doctor but becomes an important assistant in medical practice. From 2025, artificial intelligence will be actively used in the mandatory medical insurance system, including in areas such as radiography, fluorography, and computed tomography of the chest.
Read more on the topic:
AI on guard of health: an algorithm for early cancer diagnosis has been created in Russia
38% of Russians do not trust AI in hospitals: deputies propose to legislate alternative service
