Automatic Detection and Localization of Tuberculosis in Chest X-Rays
Sabitha S V1, Jeena R S2

1Sabitha S V, M.Tech student, Department of Electronics and Communication, College of Engineering Trivandrum, Kerala, India.
2Jeena R S, Assistant Professor in the Department of Electronics and Communication Engineering at College of Engineering Trivandrum, Kerala, India.

Manuscript received on 13 June 2017 | Revised Manuscript received on 20 June 2017 | Manuscript Published on 30 June 2017 | PP: 87-95 | Volume-6 Issue-5, June 2017 | Retrieval Number: E5008066517/17©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: Tuberculosis is a major health threat in many regions of the world. Opportunistic infections in immune compromised HIV/AIDS patients and multi-drug-resistant bacterial strains have exacerbated the problem, while diagnosing tuberculosis still remains a challenge. When left undiagnosed and thus untreated, mortality rates of patients with tuberculosis are high. Standard diagnostics still rely on methods developed in the last century. They are slow and often unreliable. In an effort to reduce the burden of the disease, this thesis work presents an automated approach for detecting and localizing tuberculosis in conventional postero -anterior chest raadiographs. A set of features are extracted from the lung region, which enable the X-rays to be classified as normal or abnormal using a binary classifier. Then if the chest x-ray is classified as abnormal again a set of local features are extracted to localize the affected regions . Thus it become easy to diagnose and treat the disease. An accuracy of 90% is achieved by this method.
Keywords: Graph Cut Segmentation, Classification, Local Feature Extraction.

Scope of the Article: Classification