Medical Image Analysis for TB Diagnosis System Kumari P. Prasanna1,*, Rao B. Prabhakara2 1Research Scholar, ECE Dept., JNT University: Kakinada, Kakinada, 533003, India 2Program Director, School of Nanotechnology, IST, Kakinada, 533003, India *Corresponding Author: P. Prasanna Kumari, Research Scholar, ECE Dept., JNT University: Kakinada, Kakinada, 533003, India, e-mail: prasannasudhirg@gmail.com
Online published on 4 April, 2020. Abstract Tuberculosis (TB) is a disease caused by bacteria called Mycobacterium tuberculosis. It usually spreads through the air & attacks low immune bodies. Recently, several techniques are applied to diagnosis the TB diseases. Unfortunately, diagnosing TB is still a major challenge. In recent years, a variety of techniques have been developed. In this paper, texture feature set is obtained using three different categories like statistical, structural and gray level dependent features. After that, the feature selection scheme is carried out and TB classification is done using GANN classifier. In GA-NN, genetic algorithm and neural network are combined to do the classification process. Once the abnormal TB is classified via GA-NN classifier, the TB region is identified via morphological operator. Experimental results demonstrate that the proposed method outperforms than the existing method. Top Keywords Tuberculosis, GA-NN, Kernel FCM. Top |