Lung Tuberculosis Detection Model In Thorax Radiography Image

Abstract

Tuberculosis (TB) is a communal disease with high death and disease rates worldwide. The chest radiograph (CXR) is commonly used in diagnostic solutions for lung TB. Automatic computer-aided solutions to identify TB using CXRs can advance the efficiency of the diagnostic of TB. In this study, an automatic TB detection model using CXR image is proposed. By identifying open issues from the current the state-of-the-art; how to detect the lung region automatically without any input like lung mask or model, and what are the features can identify a given CXR image is infected or normal are considered using three public datasets such as Schengen, Montgomery Country (MC), and JSRT set a total of 1047 CXR images. The proposed model used a new approach which is clear the intensities that interconnected with the border CXR image then continue to normal and further processes in image processing. The possible statistical textural features of a lung object is obtained from the first-order and second-order Gray Level co-occurrence Matrix (GLCM) statistical features. The performance of the proposed model was evaluated using accuracy, sensitivity and specificity. The model was achieved AUC 91%, 62%, 71%, and 81% on Schengen, JSRT, MC and Combined set respectively.