Abstract
The application of machine learning, particularly the *Convolutional Neural Network* (CNN) method, has experienced rapid development in the medical field. Through the process of image data grouping, CNN is capable of classifying based on the similarity of characteristics and attributes of the images. This capability is highly beneficial in the healthcare sector, especially in differentiating chest X-ray images. This study aims to evaluate the DenseNet architecture while applying it to classify Covid-19 disease by utilizing 98 chest X-ray images as training data. The test results showed that the average *loss* value on the training data was 0.8621, with an average accuracy of 0.6950. Meanwhile, the average *loss* value on the validation data was recorded at 1.0682, with an average accuracy of 0.6399. From these results, it can be concluded that the model performed better on the training data than on the validation data, indicating the occurrence of *overfitting*. Therefore, further adjustments to the model are necessary to improve its performance on the validation data