Screening of chest X-Rays for Tuberculosis using Deep Convolutional Neural Network
R.Rohith1, S.P.Syed Ibrahim2
1R.Rohith, School of Computer Science and Engineering Vellore Institute of Technology-Chennai.
2S.P.Syed Ibrahim, School of Computer Science and Engineering Vellore Institute of Technology-Chennai.
Manuscript received on August 24, 2020. | Revised Manuscript received on January 28, 2021. | Manuscript published on January 30, 2021. | PP: 254-258 | Volume-9 Issue-5, January 2021. | Retrieval Number: 100.1/ijrte.C4460099320 | DOI: 10.35940/ijrte.C4460.019521
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Tuberculosis is a life-threatening disease that mainly affects underdeveloped as well as developing nations. While lethal it is often resistive to antibiotics and the safest way to treat a patient is to detect the disease’s presence as soon as possible. Various techniques have been developed to diagnose tuberculosis and radiography of the chest is one of such methods that works well for over a decade.. Though an effective method still the success depends on the medical officer who examines the chest X-rays. Thus ,this paper proposes an approach for detecting X-ray abnormalities using deep learning. The systems output is assessed on two open Montgomery and Shenz en chest X-ray datasets and accuracy of 84 percent is achieved.
Keywords: Image augmentation, deep learning, radiography.