Comparative Analysis of Deep Convolution Neural Networks on Medical Image-Based COVID-19 Detection

The COVID-19 pandemic has spurred a dire need for efficient and accurate diagnostic tools to combat the spread of the virus. Medical imaging, particularly chest X-rays and computed tomography (CT) scans, has emerged as a vital component in the diagnosis and monitoring of COVID-19-related lung abnormalities. An effective method is presented in this paper for identifying COVID-19 and distinguishing it from normal X-rays and viral pneumonia cases using Deep Convolutional Neural Networks (CNN). An assessment of three CNN models (EfcientNetB0, VGG16, and InceptionV3) is performed using transfer learning methods. The most common deep learning algorithms used in this study are convolutional neural networks (CNNs). Performance measures (accuracy, recall, specificity, precision, and F1 scores) and deep learning approaches are used in this work. With an overall accuracy of 96.27% and a sensitivity of 95.63% for COVID-19, the findings demonstrate that the suggested technique generated a high-quality model. Effective detection and screening procedures may be facilitated by implementing computer vision design, as demonstrated by the study.