The intricate mechanism of our lungs involves frequent expansion and contraction throughout the day, facilitating the essential exchange of oxygen and carbon dioxide. Any damage to the respiratory system can lead to the development of lung diseases, which are prevalent worldwide. This category encompasses chronic obstructive pulmonary disease (COPD), asthma, tuberculosis (TB), cancer, and other prevalent conditions. Notably, lung cancer stands as a primary cause of global mortality, underscoring the critical importance of timely detection. Predominantly stemming from factors such as smoking, infections, and genetic predispositions, the majority of lung disorders pose significant health risks. Diagnostic procedures typically involve examining CT/X-Ray images of the patient's lungs, a process known for its time-consuming nature. To streamline and enhance this process, deep learning-based technologies have proven effective. In the current era of big data, traditional computational models fall short in accurately detecting lung diseases. Therefore, the integration of deep learning techniques becomes imperative, enabling the processing of image datasets for more efficient learning and predictive capabilities. In this study, Dense Net architecture is employed to classify lung diseases, and an accuracy of 86% is achieved. The simulation results demonstrate the effectiveness of the proposed model over other baseline models used for comparison.

Deep learning Lung disease detection Convolutional neural networks Dense Net