Automated Lung Disease Diagnosis Using Machine Learning And Deep Learning Techniques

Lung disorders like pneumonia, bacterial infections and tuberculosis are major health problems where the health care system faces the burden of their diagnosis and treatment. Conventional diagnostic approaches rely on manual interpretation of medical images, which requires qualified professionals to analyse. However, manual analysis is prone to human error. Therefore, this research uses deep learning and machine learning approaches to develop an automated system for lung illness identification in order to address these difficulties. In order to maximize classification accuracy, the suggested methodology combines hybrid architectures with Convolutional Neural Networks (CNNs) and pretrained RESNET 50 models for feature extraction from chest X-rays (CXRs) and also compares the with Machine learning models.. The experimental review shows that the system has the potential to improve diagnosis accuracy, streamline clinical operations, and help healthcare professionals make educated decisions using Naïve Bayes recommended for HOG features and Decision Tree for LBP features in lung disease prediction tasks. The suggested system provides scalability for wider medical imaging applications in addition to addressing present difficulties in the diagnosis of pulmonary diseases.