AI in Russia Is to Help Doctors Choose the Right Medications
Researchers in Russia’s Far East have developed an artificial intelligence system designed to support physicians in complex and non-standard drug treatment decisions.
A new AI tool has been developed by specialists at the Institute of Automation and Control Processes of the Far Eastern Branch of the Russian Academy of Sciences. The system is a hybrid software platform for planning and monitoring drug therapy, analyzing patient data, prescribed medications, and clinical scenarios to help doctors make more informed decisions.
Reasoning by Analogy
The system’s key feature is its combination of knowledge graphs and analogical reasoning. This means it relies both on formalized medical data and on patterns drawn from clinical practice.
Knowledge graphs structure information about diseases, symptoms, medications, contraindications, and drug interactions as interconnected elements.
The analogical reasoning mechanism complements this foundation by searching for similar clinical cases - situations where patients had comparable symptoms, comorbidities, or treatment histories - and analyzing which treatment regimens proved effective.
As a result, the system first identifies treatment options that are medically acceptable and then helps physicians select the most appropriate approach for a specific case, including atypical situations, while remaining within professional guidelines.
This is particularly important in personalized medicine, where standard treatment protocols do not always deliver optimal outcomes.
Thinking Like a Doctor
According to the developers, the system can operate with incomplete or even contradictory information, bringing its logic closer to the way experienced physicians think. In this model, AI does not replace the doctor but acts as an assistant that reduces the risk of prescribing errors and saves time.
Part of a Broader Trend
The Far Eastern research team’s work fits into the broader trend of digital transformation in Russia’s healthcare system, where increasing emphasis is placed on practical applications of artificial intelligence. Similar systems are already used for medical image analysis, diagnostic support, and medical record management.
The next stage for the new system will be testing in real clinical settings. This will make it possible to evaluate its effectiveness in everyday medical practice and assess how the technology could be scaled for healthcare institutions at different levels.
