Accurate TB Manifestation Using Multi Class SVM Classifier

TB is one of the leading cause of death worldwide, with a mortality rate of over 1.2 million people in [2010]. When TB is left undiagnosed, mortality rates will be high. This paper presents an accurate approach for detecting TB using a well-known classifier known as the Multiclass SVM classifier. In this paper, we first extract the lung region using a graph cut segmentation method. For this lung region, we compute a set of texture and shape features, which enables the X-rays to be classify the lung region as normal, moderate or severe (TB affected) using a Multi-class SVM Classifier. In an effort to reduce the burden of TB, this recent approach achieves a maximum accuracy in identifying TB. This proposed system for TB manifestation achieves an accuracy of 94.3% compared with the earlier methods [1] which achieves an accuracy of 86%. We collect the dataset from SKS hospital and perform the classification for the received dataset. We compare the performance of the received dataset with the classifiers: KNN, SVM& Multi-class SVM classifier. Among the classifiers, the Multiclass SVM Classifier achieves a maximum accuracy. Hence the Multi-class SVM classifier is promising in achieving the maximum performance up to the human experts.