Translation. Region: Russian Federation –
Source: Novosibirsk State University –
An important disclaimer is at the bottom of this article.
Researchers at the Novosibirsk State University's Artificial Intelligence Center have developed and patented the "Safe Medicines" program, which helps physicians select the optimal therapy based on individual patient characteristics. The development allows for the evaluation of drug compatibility, potential side effects, and the impact of medications on specific clinical and laboratory parameters.
The program is one of the key modules of the "Doctor Pirogov" medical decision support system, developed by the same team of specialists. It can also be used as a standalone service.
The relevance of such development is associated with the growing number of patients receiving combination therapy.
"Today, it's almost impossible to see a patient with just one problem. Doctors are faced with a situation where they need to prescribe five or six medications, taking into account their interactions, side effects, and the patient's individual characteristics," said Vladimir Ivanisenko, project manager, leading researcher at the NSU Center for Artificial Intelligence, and head of the computational proteomics laboratory at the Institute of Cytology and Genetics, Siberian Branch of the Russian Academy of Sciences.
Each medication has its own side effects and can have different effects on the body depending on age, comorbidities, and test results. As a result, doctors must consider dozens of factors simultaneously: existing medications, clinical manifestations of the disease, laboratory data, potential drug interactions, and the individual characteristics of the patient. Keeping all this information in mind is extremely difficult, especially under high workload conditions. Automation of such analysis helps ensure that important details are not missed and reduces the likelihood of medical error, making therapy safer and more effective.
Unlike other medication selection systems, "Safe Medicines" analyzes not only standard treatment regimens and drug interactions but also the patient's specific clinical picture—laboratory parameters, symptoms, and comorbidities. The system uses artificial intelligence methods and semantic knowledge graphs, enabling it to generate more accurate and easily interpretable recommendations for the physician.
"We select medications that don't exacerbate existing disorders and are compatible with the patient's current conditions. Such solutions are currently unavailable in Russia or anywhere else," noted Vladimir Ivanisenko.
As mentioned earlier, the development is part of the "Doctor Pirogov" project—a digital physician assistant designed to support clinical decision-making. The system integrates knowledge across 20 medical specialties and contains information on over 250 diseases. It analyzes medical records, laboratory and instrumental test results, generates a list of probable diagnoses, and offers treatment recommendations taking into account potential drug interactions.
According to the developers, the "Safe Medicines" module is particularly useful when working with complex clinical cases where standard clinical guidelines are insufficient. A private medical clinic has already expressed interest in implementing the technology, ready to integrate the service into its medical information system (MIS). NSU expects the first implementation of the program to take place in the coming months. It will also eventually be introduced into the public healthcare system, both as a module of the "Doctor Pirogov" digital physician assistant and as a standalone product. Interest in the service has already been high: consultations are currently underway with representatives of the regional Ministry of Health.
Please note: This information is raw content obtained directly from the source. It represents an accurate account of the source's assertions and does not necessarily reflect the position of MIL-OSI or its clients.