Lung (Pulmonary) diseases such as Lung Opacity and Viral Pneumonia continue to be major public health concerns, contributing significantly to global morbidity and mortality. Early and accurate diagnosis is essential for effective treatment and better patient outcomes. Chest X-ray imaging remains one of the most accessible and cost efficient tools for lung disease screening; however, manual interpretation often depends on expert radiologists and is susceptible to human error, particularly in low-resource healthcare environments. To overcome these limitations, this study proposes a deep learning-based framework for automated lung disease classification using chest X-ray images. A publicly available dataset from Mendeley Data was used, containing normal and diseased lung images. Several convolutional neural network (CNN) architectures, both custom and pretrained, were evaluated to determine their performance in automated lung disease classification. The pretrained models—ResNet50, VGG19, ImageNet227, MobileNetV3, DenseNet169, Xception, Inception, NasNetMobile, and EfficientNetV2—were fine tuned and compared against a baseline CNN model. Among these, DenseNet169 achieved the highest accuracy of 95.97%, followed by EfficientNetV2 (94.81%) and MobileNetV3 (93.37%). Experimental results show that deep transfer learning models outperform traditional CNNs, offering significant potential for clinical diagnostic support.