Computational intelligence in the identification of Covid-19 patients  by using KNN-SVM Classifier

Initiatives to mitigate the persistent coronavirus disease 2019 (COVID-19) crisis shown that quick, sensitive, and extensive screening is essential for managing the present epidemic and future pandemics. This virus seeks to infect the lungs by generating white, patchy opacities inside them. This research presents an advanced methodology employing deep learning techniques for the analysis of medical pictures pertaining to respiratory disorders. This experiment included two data sets, the initial one including normal lungs sourced from the Kaggle data pool. We acquired the anomalous lungs from https://github.com/muhammedtalo/COVID-19. We applied Principal Component Analysis (PCA) and Histogram of Gradients (HOG) as extract features. while we conducted a classification process using K nearest neighbors (KNN) and Support Vector Machine (SVM) algorithms . Results showed that the classification accuracy with SVM for Covid-19 identification is 88.54% while with KNN is 82.31%