Natural language processing (NLP) methods are used to analyze unstructured medical information, including recognition of medical terms and extraction of their relationships. Only a few medical services that use NLP methods for analyzing unstructured text are officially registered in Russia. The objective of this study was to develop a unified terminological system to enable the creation of standardized, interoperable applied solutions. NLP methods, including large language models and machine learning algorithms, are used to process the data from medical literature, real clinical practice, and current clinical guidelines. Initially, the National Unified Terminology System (NUTS) was developed, it contains the most well-known international meta-thesauruses, mapped with the terms of the federal Russian thesauri. Subsequently, services were developed based on the ontological structure of the NUTS to address the following tasks: automated knowledge bases construction for clinical decision support systems, annotation of unstructured medical text and semantic grouping of extracted terms (e.g., anamnesis, diagnosis, treatment), medical information retrieval (implemented as a chatbot for physicians, researchers, and students). The results of the work are applicable both at the level of medical organizations and at the federal level as data center-based services.