Lung diseases, such as pneumonia and lung opacity, contribute significantly to morbidity, mortality worldwide. Early detection and accurate classification of these conditions through medical imaging plays decisive role in patient outcomes. In this work, we propose a ML- based approach to classify lung X-ray images into three categories: normal, lung opacities, and viral pneumonia. We trained the model using Convolutional Neural Network (CNN) using the InceptionV3 architecture, with an accuracy of 91%. ML model was trained using an ImageDataGenerator for data augmentation and evaluated on a separate test set. Performance metrics such as accuracy, loss, and confusion matrices were used to measure the model’s effectiveness. Our results demonstrate that the proposed method achieved high accuracy in classifying lung X-ray images. Additionally, we discuss the importance of these results with reference to medical diagnostics and outline potential future work to upgrade the model’s performance. By integrating this model into a diagnostic framework, healthcare professionals can leverage it for rapid and accurate identification of lung conditions. This has the potential to assist in early intervention, reduce misdiagnoses, and ultimately contribute to better.