SELECTING TIME SERIES NESTING PARAMETERS TO DETECT DATA ANOMALIES IN BEHAVIORAL ANALYTICS SYSTEMS

The purpose of the study: the digitalization of healthcare has stimulated the development of medical information systems. Studies have become more accessible due to the widespread implementation of digital image exchange in radio logy services, as well as the increase in the number and improvement in the quality of radiological equipment. The growing load on the existing infrastructure of medical institutions requires constant administration and prompt resolution of failures. The aim of this work is to develop approaches for the effective maintenance of medical information systems. Method: the architecture of a distributed PACS/RIS system was examined. The main components of the PACS/RIS system involved in its operation were presented. Results: a developed tool for monitoring and managing medical information systems is presented. Its functional capabili ties include: managing the configuration of various DICOM objects; obtaining information about message exchange; managing application software; generating reports on their workflow stages. Reviews were conducted aimed at improving medical information systems from the perspective of information security: methods of embedding digital watermarks for radiological images; documents regulating modern requirements for technical measures and equipment to address data security threats in medical information systems. Scientific novelty: this work provides recommendations for creating digital tools that monitor and manage medical information systems, taking scalability into account.