Identification of tuberculosis bacteria based on shape and color

A report on Tuberculosis in Monkeys (Macaca mulatta): A case study at Chittagong Zoo

“…For this reason, bacilli looked red and the tissue looked blue in the acid fast staining result. In a study of Forero, M. et al, (2004), similar identification of red color TB organism was found by Ziehl-Neelsen method for acid-fast staining. (Table 2).…”

Autofocus evaluation for brightfield microscopy pathology

“…8 In this study, a wide set of focus measures from the already well known methods to those proposed recently have been analyzed. Some of these measures have been specifically proposed for autofocusing bacteria specimens, 9,10 while others have not been tested within this particular context. 11 Other focus measures, such as 4 Brenner gradient and entropy method, 3 have not been included here, but they belong to the same family of Vollath and histogram techniques.…”

Image Segmentation of Ziehl-Neelsen Sputum Slide Images for Tubercle Bacilli Detection

“…From the resulted images that have been presented, it can be said that this technique could be an alternative solution for the image segmentation of TB bacilli, and to further helps the process of TB bacilli identification as well as classification in sputum samples. Most of the research that have been done previously used fluorescence images of sputum (Veropoulos et al, 1998;Veropoulos et al, 1999;Forero et al, 2003;Forero et al, 2004), whereby this research concentrates on the Ziehl-Neelsen stained images of sputum. Hence, it provides another option of TB bacilli identification especially for developing countries which are still sticking to this method for TB detection.…”

“…This logically equivalent to segmenting each image plane individually, creating separate binary images and then combining them with a Boolean AND operator afterward. This color thresholding method is widely used in the image segmentation (Forero et al, 2003;Forero et al, 2004;Mancas-Thillou & Gosselin, 2005).…”

“…Some image processing algorithms also have been developed in order to carry out automatic TB bacilli detection in the captured image. Forero applied adaptive color thresholding technique to the images that have been captured using fluorescence microscopy (Forero et al, 2003;Forero et al, 2004). Veropoulos used an identification method based on shape descriptors and neural network classifiers (Veropoulus et al, 1998;Veropoulus et al, 1999).…”