@article { author = {Taghinezhad, F and Kaffashian, M and Kalvandi, Gh and Shafiei, E}, title = {Explanation of COVID-19 Mortality Using Artificial Neural Network Based on Underlying and Laboratory Risk Factors in Ilam, Iran}, journal = {Archives of Razi Institute}, volume = {77}, number = {3}, pages = {1327-1332}, year = {2022}, publisher = {Razi Vaccine & Serum Research Institute}, issn = {0365-3439}, eissn = {2008-9872}, doi = {10.22092/ari.2021.355438.1684}, abstract = {The spread of new waves of coronavirus outbreaks, high mortality rates, and time-consuming and numerous challenges in achieving collective safety through vaccination and the need to prioritize the allocation of vaccines to the general population have led to the continued identification of risk factors associated with mortality in patients through innovative strategies and new statistical models. In this study, an artificial neural network (ANN) model was used to predict morbidity in patients with coronavirus disease 2019 (COVID-19). Data of 2,206 patients were extracted from the registry program of Shahid Mostafa Khomeini Hospital in Ilam, Iran, and were randomly analyzed in two training (1,544) and testing (662) groups. By fitting different models of a three-layer neural network, 12 variables could explain more than 77% of the mortality variance in COVID-19 patients. These findings could be used to better mortality management, vaccination prioritization, public education, and quarantine, and allocation of intensive care beds to reduce COVID-19 mortality. The results also confirmed the power of a better explanation of ANN models to predict the mortality of patients.}, keywords = {Artificial Neural Networks,covid-19,Multilayer perceptron,Iran}, url = {https://archrazi.areeo.ac.ir/article_126569.html}, eprint = {https://archrazi.areeo.ac.ir/article_126569_d490e61df3d447b52d2f6072aebe2e5d.pdf} }