INTELLIGENT INFORMATION TECHNOLOGIES FOR MODELING PROCESSES IN THE HEALTHCARE SYSTEM

Abstract

The article examines the application of artificial intelligence (AI) technologies and data mining approaches in the healthcare sector. The study focuses on the integration of data mining methods, machine learning, and deep learning for identifying patterns within large volumes of medical information. A comparative analysis of the most widely used Data Mining models in medicine—including classification, prediction, clustering, and hybrid architectures—is conducted. Their advantages and limitations are assessed in terms of predictive accuracy, adaptability to changing conditions, and interpretability of results. The article analyzes algorithms that have demonstrated high effectiveness in tasks such as diagnostics, pathology detection, medical image processing, and disease risk prediction. It is shown that AI methods and models are applied not only in medical diagnostics but also in supporting managerial decision-making for efficient hospital resource management, optimization of administrative and economic processes, and risk assessment. The key challenges related to data quality, system integration, and the lack of standardized model evaluation criteria are identified. The study concludes that a multimodal approach enhances prediction accuracy, model generalizability, and decision explainability, which is essential both for clinicians and patients and for improving economic and managerial processes in healthcare institutions. The research substantiates the need to develop adaptive and explainable models based on Explainable AI (XAI) to ensure greater transparency, trust, and interpretability of predictions. Based on an analysis of modern scientific approaches, a conceptual model of an intelligent decision support system is proposed, incorporating sequential stages of data collection, cleaning, preprocessing, integration, and analytical processing of medical data. The proposed model is oriented toward improving the accuracy of diagnostic decisions, optimizing resource allocation in healthcare facilities, and ensuring the explainability of outcomes. The results demonstrate that integrating AI technologies into the healthcare system reduces diagnostic errors, advances personalized medicine, and supports the formation of predictive treatment strategies using real-time data analysis. The proposed approach can serve as a foundation for developing effective digital healthcare solutions and national AI-driven clinical decision support systems.