A multi-agent system for modeling tuberculosis transmission

Abstract

Introduction. Forecasting epidemiological processes holds immense importance as it allows for understanding and anticipating future disease and epidemic trends. The aim of the study was the development of a multi-agent system for simulating the transmission of tuberculosis infection. Materials and methods. The primaiy aim of this study was to develop a model that accurately simulates the transmission of tuberculosis within an urban setting. The modelling process itself is characterized by a series of key stages, including initialization of the city, calibration of health parameters, simulation of the working day, propagation of the spread of infection, the evolution of disease trajectories, rigorous statistical calculations, and transition to the following day. Results. The model’s results exhibit stability and lack of significant fluctuations. The statistical values obtained for infected, latent, and recovered individuals align well with known medical data, confirming the model’s adequacy. The simulation time for a model with 100,000 agents is approximately 30 minutes, enabling parallelization of processes for modeling multiple cities, regions, or countries. This opens the possibility of using computer clusters and optimizing TB prevention strategies based on reinforcement learning neural networks. The proposed model allows for not only statistical data but also individual-level analysis of the tuberculosis spread by specific agents. Conclusion. The proposed model allows for tracking and analyzing the life and behavior of each individual agent, enabling a thorough assessment of tuberculosis infection spread and the development of prevention strategies.