TUBERCULOSIS DETECTION USING SVM AND CNN
Tuberculosis is the deadly disease caused by Mycobacterium Tuberculosis(MTB) that generally affects the lungs and other part of the body. Many patients are died due to lack of treatment and error in diagnosis. Chest X-ray(CXR) image has been the main tool for detecting lung TB historically. Nowadays lots and lots of machine learning tools in Bio-medical field for detecting this disease. In this paper, there are two methods of detecting tuberculosis that is Support vector machine (SVM) and Convolutional Neural Network (CNN). The use of deep learning technique is because of its cheapness of these imaging techniques. Tuberculosis detection using deep learning can classifies the X-ray images into abnormal and normal which is the emerging technique for TB surveillance. In this paper various machine learning technique are used that aims in improving the accuracy of tuberculosis prediction has been applied. In tuberculosis detection , it is very important to find the accurate predictions. But, the traditional methods approaches like MODS which take long time to detect the tuberculosis and that also with inadequate prediction and diagnosis. So the dataset that is the CXR images are taken from the laboratories is utilized with Support Vector Machine as well as with deep learning model. The result shows the improvement in the classification accuracy. This paper can train the specialized architecture from scratch and achieve good results and also increase the accuracy with large amount of datasets.