INTEGRATED HYBRID MODEL FOR LUNG DISEASE DETECTION THROUGH DEEP LEARNING

The burden of lung diseases on world health is substantial, underscoring the vital necessity of timely detection. The VGG16 architecture with additional convolutional layers is used in this study to provide a hybrid method to lung disease classification. It incorporates the Synthetic Minority Over-sampling Technique (SMOTE) to improve model performance in response to the problem of imbalanced class instances. The subset of the NIH Chest X-ray dataset is used to train and assess the model. The designed model classifies the images into 8 different classes of lung diseases. They are Emphysema, Cardiomegaly, Atelectasis, Edema, Consolidation, Mass, Effusion, Pneumothorax. The proposed model delivered accuracy of 96.42% which demonstrates the efficacy in precise classification of lung diseases. The Graphical User Interface (GUI) is integrated for better interaction between the patient and the model. Through improved diagnostic capabilities, this suggested method not only promotes technological innovation but also shows promise for enhancing patient care and health care outcomes.