Lung diseases encompass a broad spectrum of respiratory disorders affecting the lungs, including infections, chronic conditions like asthma or COPD, and more severe issues such as lung cancer. These diseases often impact breathing function and necessitate diverse medical interventions for diagnosis, management, and treatment. The presented work proposes a comprehensive approach using chest X-ray images based on deep learning for accurate disease identification in medical imaging. The study initiates with the integration of raw chest X- ray images into various deep learning models, including MobileNet, MobileNetV2, Inception V3, ResNet-50, and VGG-16. Among these models, MobileNet emerges as the most proficient performer with an accuracy of 91.7%. Subsequently, the investigation expands into two primary aspects. In the first phase, the application of image enhancement techniques, namely Contrast Limited Adaptive Histogram Equalization (CLAHE) and Black Hat Transformation, is explored. The processed data are then fed into MobileNet to evaluate their impact on classification results. In the second phase, the study examines into fine-tuning MobileNet using preprocessed data to observe any further improvements in classification outcomes with an accuracy of 96% and AUC score of 0.99. The proposed methodology confines a multi-step process, involving model selection, image enhancement, and fine-tuning, with a focus on optimizing the deep neural network’s performance for enhanced lung disease classification accuracy.