A Novel Machine Learning Approach for Tuberculosis Segmentation and Prediction Using Chest-X-Ray (CXR) Images

Tuberculosis is a potential fatal disease with high morbidity and mortality rates. death rates are rising, posing serious health threat in several poor countries around the world. To address this issue, we proposed novel method for detecting tuberculosis chest X-ray (CXR) images that uses three-phased approach to distinguish such as segmentation, feature extraction, classification. In CXR, utilized Weiner filter reduce impulse noise. The features were extracted from CXR trained using decision tree classifier known stacked loopy (SLDT) classifier. For classification process, ROI-based morphological was applied mentioned approach, extraction accomplished through chromatic Prewitt-edge highlights.