COVID-19 Classification using CNN, Inception V3 and Transfer Learning

The outbreak of the COVID-19 pandemic has posed unprecedented challenges for global healthcare systems, neces-sitating accurate and rapid diagnostic solutions. The suggested method utilizes the capabilities of Convolutional Neural Networks (CNN) and the Inception V3 module to provide a novel deep learning strategy for COVID-19 identification. The capacity of the Inception V3 architecture to extract nuanced features from challenging medical imaging data is famous. This work evaluates the performance of the CNN-Inception V3 model in differentiating COVID-19 instances from other respiratory illnesses, such as pneumonia, fibrosis, tuberculosis, and normal lung scans, through a thorough examination employing a broad dataset of chest X-ray and CT images. Important parameters like accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC-ROC) are used to assess the model’s performance. The study also explores the interpretability of CNN-Inception V3 model predictions, providing information on possible clinical uses. The results of this study add to the corpus of research on deep learning-based COVID-19 diagnostics and show the value of cutting-edge technology in the fight against infectious diseases.