Lungs Disease Detection Using Convolutional Neural Networks and Guided Image Filtering

In the recent past, due to the Covid-19 emergency many people died. Therefore, to save life accurate and timely detection of lung diseases, like pneumonia and COVID-19, is critical for effective treatment of the patient. This study presents a simple methodology for lung disease detection which combines the Convolutional Neural Networks (CNNs) with Guided Image Filtering (GIF). The integration of GIF with CNNs aims to enhance image quality by reducing noise and improving edges in the lungs image, which facilitating more precise feature extraction and classification. The experiment is conducted on the extensive database. Experimental results demonstrate that the CNN-GIF model significantly outperforms standard CNN approaches, achieving higher accuracy. The comparative analysis highlights the potential of this advanced technique in improving the accuracy of the diagnosis and helping the healthcare specialist in making correct decisions. The proposed approach is compared with traditional and state of the art methods to assess its efficacy in distinguishing between pneumonia, COVID-19, and normal lung cases.