A Comparative Study for Chest X-rays Indicating Tuberculosis

With over 10 million infections and 1.5 million fatalities recorded from Tuberculosis (TB) in 2019 alone, it is the top cause of death worldwide. For this condition to be effectively treated, precise and prompt TB identification is essential. An imaging technique that is frequently used to diagnose TB is the Chest X-ray (CXR). This study suggests a deep learning-based method for CXR-based TB detection to overcome these difficulties. On a dataset of CXRs from the dataset that included both normal and TB-infected images it assesses the proposed approach and achieves an accuracy of 83.33% in TB detection. Then, we applied Singular Value Decomposition (SVD) and Principal Component Analysis (PCA). It is possible to develop more precise and trustworthy diagnostic methods for detecting TB by utilizing PCA and SVD to extract and analyze features from CXRs, which will ultimately assist in improving patient outcomes and stopping the spread of the illness.