Treffer: Modeling the Propagation of COVID-19 Using a Multilayer Perceptron and Radial Basis Function in Digital Twins Framework.

Coronavirus-19 (COVID-19), as the most recent global pandemic, has significantly affected individuals' daily habits and mobility. These alterations have been related to location, and they can be predicted using Geospatial Artificial Intelligence (GeoAI) modeling. GeoAI and geovisualization serve as essential tools for gaining clearer insights into the application of spatial phenomena in reality. Digital Twin (DT) as a visualization technology combines software and human efforts in research, particularly in healthcare. They create virtual replicas of patients for disease modeling, allowing for personalized medicine by simulating disease progression and treatment responses. This enhances predictive accuracy and helps develop tailored therapeutic strategies. This paper aims to detect spatial patterns and effective criteria in the outbreak of COVID-19 using GeoAI within the framework of DT. The main contribution is the application of kernel-based algorithms to the disease distribution pattern. The applied data is organized into three general categories: infrastructure (distance to road, land use), environment (traffic congestion, air pollution), and socioeconomic (population density, gender ratio, income, education level). Each of these categories has its own sub-criteria. The Multilayer Perceptron (MLP) considers the relationships of input targets based on a normal distribution, while the Radial Basis Function (RBF) technique considers the assumption of a radial influence zone. The COVID-19 dataset was collected over four months from eight hospitals in Tehran. The interpretation of the results indicates that the RBF network, with an RMSE of 1.77e-08, models the COVID-19 outbreak more accurately than the MLP, which had an RMSE of 0.0037. The application of DT with MLP and RBF represents a powerful approach to modeling and simulating complex systems. Utilizing the Artificial Neural Network (ANN) algorithm within the digital twin framework, health centers can achieve enhanced predictive capabilities and real-time responsiveness, improving treatment processes across various medical domains. [ABSTRACT FROM AUTHOR]

Copyright of Earth Observation & Geomatics Engineering is the property of University of Tehran and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)