Medical image analysis is crucial for the early detection of lung illness, assisting doctors in providing appropriate therapies, and preventing fatalities. In this paper, an automated system is developed by fusing metaheuristic algorithm and machine learning classifier to differentiate between heathy lungs and affected lungs. The developed system has four phases such as preprocessing, feature extraction, feature selection, and classification. In the first phase, affected region is isolated from its background using thresholding method. The segmented image is used to construct the second phase's Gabor features and Gray level co-occurrence matrix. Dimension of the feature vectors are reduced with principle component analysis in third phase. In order to categorize the images, Support Vector Machine (SVM) is employed. Due to their increased accuracy, experimental findings demonstrated that the suggested approach is acceptable for identifying lung diseases and may be used in real-time.