Performance Evaluation of Ensemble Deep Learning Algorithms for Prediction of Pandemic Disease

For optimal healthcare management and counter-measures, it is essential to monitor and predict severe disease at the right time, before it becomes pandemic. In this research work, the most recent pandemic is considered as an example, as the viral coronavirus (COVID-19) prognosis is crucial to learn from. The severe COVID-19 threat has had a substantial influence on the global health security scene, forcing the creation of cutting-edge computer models to imorive monitoring, control, and mitigation measures. The research study aims to develop a generalized model assessing the healthcare parameters at a personalized and community dimensions and predicting the severity of the disease before becoming pandemic. To achieve this aim, this paper has systematically evaluated the outcomes of different experiments utilizing the ResNet, DenseNet, and ensemble models using a variety of performance criteria. The ensemble model consistently demonstrated superior performance across all metrics, exhibiting an accuracy and f1-score of 97%. In comparison, the DenseNet model earned an accurancy and f1-score of 93%, while the ResNet model earned an accurancy and f1-score of 88%. All models in this paper demonstrated promising accuracy and the potential to ain in COVID-19 prediction. Chest x-ray images were employed to experiment the computational models of accurately predicting the disease. Such experiment allows us to have a better understanding of the advantages and disadvantages of various computer models for predicting sever disease, which will help create more precise and effective prediction systems fr medical condition. The achieve result highlights the efficacy of ensemble techniques for exploiting the synergistic benefits of multiple models. The knowledge gained from this study aims to go beyond the theoretical sphere and expand its influence into the real world of hospital administration.