Artificial intelligence in healthcare system. Clinical guides.

Introduction. The growing interest in the use of artificial intelligence (AI) technologies in healthcare is driven by IT industry advancements and the need to improve the quality of medical care. In the Russian Federation, from 2025, healthcare delivery must comply with clinical guidelines (CGs), which regulate the inclusion of medical interventions that have proven efficacy and safety. The inclusion of software medical devices (SMDs) with AI in CGs is complicated by their high variability, which necessitates the adaptation of existing assessment criteria. Methods. A quantitative and qualitative analysis of regulatory legal acts, instructions for the use of SMDs with AI, scientific publications, and approved CG texts was performed. Data from the State Register of Medical Devices and product instructions were studied to determine their clinical application and evidence base. Results. It was found that 93 CGs could potentially include SMDs with AI. However, instructions and publications often lack sufficient information for such inclusion: specific areas of application, device identification, and clinical effects are not always indicated. This hinders the integration of SMDs with AI into CGs. Research limitations. The analysis was limited to publicly available data from registers, product instructions, and publications, which may not fully reflect the range of SMD use in clinical practice. Conclusion. The inclusion of SMDs with AI in CGs is possible, provided that the evidence base is improved. The authors propose an adapted evidence grading scale for the evaluation of AI technologies.