AUTOMATIC SEGMENTATION OF MYCOBACTERIUM TUBERCULOSIS IN ZIEHL-NEELSEN SPUTUM SLIDE IMAGES USING SUPPORT VECTOR MACHINES

摘要

The World Health Organization suggests visual examination of stained sputum smear samples as a preliminary and basic diagnostic technique of tuberculosis disease. The visual examination requires laboratory technicians to spend considerable time, so it increases laboratorians' workload. In addition, it leads to a misdiagnosis because of requiring mental concentration. This paper presents a novel method for segmentation of tuberculosis bacteria in microscopic images taken from the Ziehl-Neelsen stained samples. Color information of bacterial regions which is taken from pixels and their adjacent pixels is sampled in training process. Multidimensional Gaussian probability density function and support vector machines are used during microscopic image segmentation comparatively. The performance of the implemented system is evaluated using sensitivity, specificity and accuracy criteria.