Ensemble Genetic Algorithms For Blood Covid-19 Diagnosis

Abstract

The state of the world is urgent at this time. The recently discovered corona virus causes Covid-19, a devastating pandemic disease. To date, there are 3.23 million ongoing cases in India, with 59,449 fatalities. There is a lot of effort being put in by scientists and doctors to
find a cure and vaccination for this. Medical advancements are being aided by research in the areas of machine learning and artificial intelligence, which are being used to forecast the development of disease and to detect the existence of the virus in the human body. The researchers here hope to learn more about COVID-19 by examining the virus in human blood samples. The study's blood samples have more than a hundred different characteristics. Therefore, the genetic algorithm has been used for feature reduction in high-dimensional data processing. This research will use a genetic algorithm to predict the presence of COVID-19 in a blood sample. About 5,644 patients' records with 111 different characteristics are included in the sample. The dimensionality reduction technique will be
based on a genetic algorithm, similar to that utilised in the optimization algorithm for ant colonies for disaster relief. The programming language python is used throughout this study, and the metrics sensitivity, specificity, accuracy, and area under the curve (AUC) are
used to assess the model's efficacy. The applied model has a 92% AUC, a sensitivity of 96.76 percent, a specificity of 98.80 percent, and an accuracy of 98.7 percent. The results showed that the custom algorithm was superior to the best existing solutions