Computer aided tuberculosis. (2021)

• Record Type: Journal Article
• Authors: Subashini, V.; Ganesan, R.; Baskaran, K.; Grace Vimala, Annie; Manikandan, ; SenthilKumar, A.
• Is Part Of: Materials today. Volume 47:Part 1(2021)
• Page Start: 292
• Page End: 298
• Publication Date: 2021
• Subjects: Tuberculosis -- Computer-aided diagnosis -- Support vector machine -- Histogram of gradients -- Active learning; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2021.04.467
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: Tuberculosis (TB) is a serious health risk in many parts of the world, and tuberculosis diagnosing remains a challenge. Undiagnosed tuberculosis patients have a high mortality rate. Modern diagnosed methods fails during mass screening for large population. It can be improved by a CAD system (Computer-Aided diagnosis) that highlights the potential areas that contains the bacterial cells. We introduce a new automated approach for TB detection which improves the diagnostic process. Chest Radiograph X-Rays are chosen as input for the TB detection. Graph Cut Algorithm is utilized for segmentation process which is the efficient methods for segmenting the affected area. Integral HOG (Histogram of Gradients) is used for featureextraction. The extracted features are classified using commingle of Support Vector Machine (SVM) and Active Learning (AL).