The quick accurate diagnosis of lung CT (L-CT), had much importance in the diagnosis of COVID-19(CoD-19). This study focuses on developing an automated system to detect CoD-19 in chest CT using the Xception model. The primary objective is to evaluate the model ability to classify CoD-19 cases and non- CoD-19 cases. A L-CT dataset was used along with data preprocessing and augmentation techniques to increase the diversity of the dataset. The proposed model, used for CoD-19 detection uses transfer learning techniques with imagenet weights and the model is evaluated by various performance measures. The model shows good results as compared to traditional techniques (accuracy = 95.6%, sensitivity = 94.8%, specificity = 96.2%, and F1 score = 95.2%). These results highlight the potential of proposed approaches to automate and improve CoD-19 diagnosis. In summary, the proposed framework demonstrates a reliable and effective diagnostic tool to help healthcare professionals make timely and accurate decisions during an epidemic and further extended to its clinical applicability.