Nowadays medicine needs high information technologies. Automation of the processes of diagnosis, stratification and treatment of diseases belongs to the priority tasks in modern medicine, since it allows increasing the efficiency and reliability of working with information, data accuracy and completeness at all stages of the medical process. Smart ward is an integral approach to automation of this process. This paper proposes an approach to creating a module for diagnosing, stratifying and selecting treatment of the paroxysmal sympathetic hyperactivity syndrome and highlights the advantages of using this module as an intelligent component of the smart ward. A data model defining medical data for this particular task is built and described using UML class diagrams. It can be seen that the model entities and their properties partially correspond to the entities and properties in the smart ward model, which allows linking local and global databases and provide higher completeness and reliability of patient information, which, in turn, will increase the efficiency of the system in general. Further research includes building a behavioral model of a smart ward and defining the scenarios of its operation, taking into account specific tasks, such as identifying paroxysmal sympathetic hyperactivity or other syndromes, determining the set of ward components necessary for the implementation of these processes, and choosing protocols for their interaction.