Detection & Extraction of Tuberculosis from Chest CT Slices

Author(s):

P. Kannan , Easwari Engineering College - Chennai; Anita Titus, Easwari Engineering College

Keywords:

Computer Aided Diagnosis (CAD), Tuberculosis, Computed Tomography, Segmentation, Classification

Abstract:

Tuberculosis (TB) is a deadly infectious disease that mostly affects the lungs, but can also affect the other organs, including central nervous system. The basic key to control the spread of TB is the early detection and treatment. A computer aided diagnosis (CAD) system is proposed to detect and classify TB. The chest CT slices are initially preprocessed to remove the gaussian noise followed by lung segmentation and then with the features extracted from the ROIs the Bayesian classifier is been trained and finally the slices are classified. A set of 30 CT(computed tomography) slices were processed that demonstrates the accuracy and satisfactory performance of the algorithm. The CAD system achieved an accuracy of 73.3%, sensitivity of 88.6% and specificity of 78.33%.

Other Details:

Manuscript Id	:	IJSTEV1I11061
Published in	:	Volume : 1, Issue : 11
Publication Date	:	01/06/2015
Page(s)	:	289-292

Download Article