Application of Artificial Intelligence for Predicting the Risk of Infectious Diseases Based on Hygienic Environmental Factors

Traditional epidemiological models are often unable to promptly and adequately account for the complex non-linear interactions between multiple hygienic factors and infectious morbidity. This work substantiates the feasibility and appropriateness of integrating artificial intelligence methods into the system of social-hygienic monitoring. Based on an analysis of contemporary literature, priority predictors of infectious risk have been identified, including indicators of drinking water quality, ambient air quality, meteorological conditions, and sanitary-living characteristics of territories. It is demonstrated that neural network algorithms can improve the accuracy of acute intestinal infection incidence forecasts by an average of 18–22 % compared to classical regression approaches. A conceptual architecture of a hybrid predictive system adaptable to the conditions of regional hygienic control is described. The conclusion is drawn regarding the necessity of phased implementation of machine learning technologies into the practice of the sanitary-epidemiological service of the Republic of Uzbekistan.