Discrete Wavelet Transforms for Lung Pneumonia Detection Using MATLAB and Image Processing

Abstract

This research paper investigates the application of Discrete Wavelet Transforms (DWT) for the detection of pneumonia using MATLAB and image processing techniques. Pneumonia diagnosis, a critical aspect of medical imaging, presents challenges in accuracy and reliability. The proposed methodology involves pre-processing lung pneumonia images, performing DWT decomposition to extract frequency components, and utilizing these components for feature extraction and subsequent classification. Classification models, including Support Vector Machines (SVM), Random Forests, and Neural Networks, are trained on extracted features to distinguish between pneumonia and non-pneumonia cases. Experimental results demonstrate the potential of DWT-based features in enhancing pneumonia detection accuracy. This paper contributes to the advancement of medical image analysis and highlights the significance of wavelet transforms in addressing complex diagnostic tasks [1].