In this paper the initialization of the city is considered, which consists of several steps, including the creation of city objects with their locations, creation of residents with their attributes and own daily schedules, etc. A description of the model is provided as a tuple of attributes. The adequacy of the simulation model is checked based on the statistical data from the city of Lviv, Ukraine. Generated locations of city ecosystem objects are presented. The daily schedule of residents is simulated. A possible work schedule for each specialty is given, and separate schedules are created for working days and holidays. A unique schedule is predicted for the resident, which depends on their age and work specialty. The dynamics of visits to facilities by residents on weekdays and at weekends are analyzed. Based on the conducted experiments, the adequacy of the model and its realistic reflection of the functioning of the city's ecosystem during the day are proven. It means that by using this model, researchers can assess the impact of different behavioral scenarios on the residents within the city ecosystem more reliably. This enables a better understanding of how certain actions or changes in behavior can affect the spread and control of diseases in a specific geographic area. This model has the potential to serve as a foundation for future modeling of systems at the medium and macro scales.

Introduction: The paper presents epidemiological process modeling, with a focus on tuberculosis utilizing multiagent system. Material and methods: This study involves the development of an algorithm that harnesses the potential of artificial intelligence to create a geospatial model that highlights the different pathways of TB transmission. The modeling 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. A comprehensive description of the course of active tuberculosis is presented, following the official hypothesis recommended by the World Health Organization. A comprehensive simulation, illustrating the propagation of tuberculosis in an entirely healthy environment devoid of any preventive or therapeutic measures, is presented. To ascertain the adequacy of the model and its sensitivity to the principal parameters governing the course of tuberculosis, a series of experiments were meticulously conducted, employing three distinct approximations, namely: the basic model, the model incorporating mortality factors, and the comprehensive model, encompassing all relevant aspects. Conclusions: 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 paper proposed a modification of the GeoSER(D) model previously developed by us by detailing the age structure of the population, personal schedule on weekdays and working days, and individual health characteristics of the agents, this made it possible to build a more realistic model of the functioning of the city and its residents. The developed model made it possible to simulate the spread of 3 types of the strain of the COVID-19 virus, and to analyze the adequacy of this model in the case of unhindered spread of the virus among city residents. The paper showed that SARS COV 2 spreads mainly from contacts in workplaces and transport, and schoolchildren and preschool children are the consequence, not the initiator of the epidemic. Fluctuations in the dynamics of various indicators of the spread of SARS COV 2 associated with the difference in the daily schedule on weekdays and weekends. It has been shown that people's daily schedules strongly influence the spread of SARS COV 2. For the more contagious "rapid" strains of SARS COV 2 (omicron), immunocompetent people become a significant source of infection. For the less contagious "slow strains" (alpha) of SARS COV 2, the most active source of infection is immunocompromised individuals (pregnant women). The more contagious – "fast" strain of the SARS COV 2 virus (omicron) spreads faster in public transport. For less contagious – "slow" strains of the virus (alpha), the greatest infection occurs due to work and educational contacts.

This study proposes a modification of the GeoCity model previously developed by the authors, detailing the age structure of the population, personal schedule on weekdays and working days, and individual health characteristics of the agents. This made it possible to build a more realistic model of the functioning of the city and its residents. The developed model made it possible to simulate the spread of three types of strain of the COVID-19 virus, and to analyze the adequacy of this model in the case of unhindered spread of the virus among city residents. Calculations based on the proposed model show that SARS-CoV 2 spreads mainly from contacts in workplaces and transport, and schoolchildren and preschool children are the recipients, not the initiators of the epidemic. The simulations showed that fluctuations in the dynamics of various indicators of the spread of SARS-CoV 2 were associated with the difference in the daily schedule on weekdays and weekends. The results of the calculations showed that the daily schedules of people strongly influence the spread of SARS-CoV 2. Under assumptions of the model, the results show that for the more contagious “rapid” strains of SARS-CoV 2 (omicron), immunocompetent people become a significant source of infection. For the less contagious “slow strains” (alpha) of SARS-CoV 2, the most active source of infection is immunocompromised individuals (pregnant women). The more contagious, or “fast” strain of the SARS-CoV 2 virus (omicron), spreads faster in public transport. For less contagious, or “slow” strains of the virus (alpha), the greatest infection occurs due to work and educational contacts.

CT scans vary from manufacturer to manufacturer. There are several generations of CT. The concept of CT generation is not so much related to the time of its birth, but to the use in the scanner of the type of motion of the emitter-detector system. The generation number (first, second, third, fourth, etc.) is related to the type of tube-detector system design. Three types of scanners are now available: third, fourth and fifth generation. The life of the first generation of CT was fleeting: the duration of the scan did not suit – about 20 minutes. and the computer reconstruction of the image took too long. Only the brain could be examined on such devices. As for the human body, the image turned out blurry, blurred. But the process of improvement continues. Today, research is being conducted on fourth- and fifth-generation devices. They have not one, but many X-ray emitters, up to 200. It uses a very powerful processor with high-speed information processing – up to 10 million operations per second. As a result, the scan time was reduced to 40-50 milliseconds. It is now possible (here is a scientific idea!) To see on the telemonitor the reduction of certain transverse layers of the heart 1-2 millimeters thick. At spiral CT in the course of research with simultaneous constant rotation of system “tube – detectors” the table constantly moves and thus there is a spiral movement of a fan-shaped beam through a patient’s body. Spiral CT makes it possible to examine the anatomical region in one period of respiratory arrest, and the thickness of the reconstructed layer is not related to the original width of the tomogram. Obtaining thin conjunctival sections, which are tightly arranged in a spiral, allows you to get three-dimensional reconstructions. In combination with intravenous bolus contrast and subtraction data processing, CT angiograms can be obtained, which reproduce images of large vessels. In recent years, multispiral (multislice) CT has been used, which is based on the principles of imaging as in spiral CT, but due to multi-row detectors for full rotation of the system “tube detectors” can display more than one slice (currently from 2 to 16 images). This has made it possible to perform studies of the heart, study a large anatomical area, such as the lungs, thin sections in one breath hold, and significantly improve the quality of multiplanar and three-dimensional reconstructions. In electron beam tomography, the source of X-rays is a large electron gun with several massive parallel anode targets located in semiconductor rings around the patient. The intense anode beam is directed along the tungsten anode rings. The electron beam thus obtained passes through the patient’s body and is received by a fixed array of detectors. This achieves a high tomography rate, which is sufficient for the study of a moving heart: clearly delineated images can be obtained without the use of synchronization with the ECG. With the help of electron-beam tomography it is also possible to perform CT angiography and construction of high-quality three-dimensional reconstructions. Thus, modern medicine is impossible without CT. For example, in neurology, computed tomography, in addition to skull bones, well identifies the ventricles of the brain, venous sinuses, subarachnoid cisterns, interhemispheric fissure, cortex, basal nodes, inner capsule, trunk, cerebellum and other structures. The radiation load does not exceed the level usual for X-ray examination. At the same time, CT scans provide 100 times more information about the brain than conventional skull x-rays. But its importance is not limited to use only for the diagnosis of various diseases. Under the control of CT perform punctures and directed biopsy of various organs and pathological foci. The role of CT in the control of conservative and surgical treatment of patients is extremely important. CT is a valuable method of accurate localization of tumors and directing the radiation source to the focus when planning radiation treatment of malignant tumors.

For machine-learning analysis, a dataset consisting of 226 observations with 68 features obtained from 257 COVID-19 patients was used. To identify the most effective models for predicting COVID-19 severity, an extensive evaluation of available classifiers from the Scikit-learn library with default hyperparameter settings was conducted, including lo-gistic regression, k-nearest neighbors, decision trees, ensemble algorithms (random forest, gradient boosting, bagging), naïve Bayes classifiers, support vector machines, and others. Model performance was assessed using Accuracy and AUC-ROC, with a focus on max-imizing AUC-ROC to ensure optimal class discrimination. Optimal hyperparameters for each model were defined as those yielding the highest mean AUC-ROC value during 5-fold cross-validation. Ensemble methods such as ExtraTreesClassifier (Accuracy: 0.974 ± 0.022) and RandomForestClassifier (Accuracy: 0.960 ± 0.035) demonstrated ROC curves approaching that of an ideal classifier (upper-left corner of the plot), confirming their ex-cellent performance in predicting COVID-19 severity. Simpler models, including Ber-noulliNB (Accuracy: 0.956 ± 0.037) and DecisionTreeClassifier (Accuracy: 0.938 ± 0.043), also showed high classification quality. Analysis of misclassification patterns revealed that ExtraTreesClassifier, HistGradientBoostingClassifier, BaggingClassifier, and Gradi-entBoostingClassifier made no errors across any class. The poorest performance was ob-served with LinearDiscriminantAnalysis, which generated 11 misclassifications, followed by CalibratedClassifierCV, and LogisticRegressionCV.