Comparison of Various Segmentation Algorithms for Automatic Tuberculosis Screening using CHEST Radiographs

Tuberculosis is a major global health problem with about 1/3rd of the world population are being affected. We try to propose an automated system for the classification of lung nodules with trusted accuracies using segmentation algorithms like quick shift clustering,SIFT,MR8-LBP,HOG using Chest Radiographs. The abstracted dataset is then classifed using classifiers like SVM and PLSA for abnormality for various algorithms. The proposed method is based on contextual analysis by combining the lung nodule and surrounding anatomical structures, and has three main stages: an adaptive patch-based division is used to construct concentric multilevel partition; then, a new feature set is designed to incorporate intensity, texture, and gradient information for image patch feature description, and then a contextual latent semantic analysis-based classifier is designed to calculate the probabilistic estimations for the relevant images. Finally, the accuracy rate of each algorithm is compared to identify the efficient one.