Diagnosis and counting of tuberculosis bacilli using digital image processing
Tuberculosis (TB) diagnosis by manual observation is inefficient because of it is time consuming and efficiency depends upon skill of the pathologist and the quality of the smear. To overcome this problem diagnosis of TB from ZN-stained sputum smear images using digital image processing is done in t...
Saved in:
Published in | 2017 International Conference on Information, Communication, Instrumentation and Control (ICICIC) pp. 1 - 5 |
---|---|
Main Authors | , |
Format | Conference Proceeding |
Language | English |
Published |
IEEE
01.08.2017
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Tuberculosis (TB) diagnosis by manual observation is inefficient because of it is time consuming and efficiency depends upon skill of the pathologist and the quality of the smear. To overcome this problem diagnosis of TB from ZN-stained sputum smear images using digital image processing is done in this paper. MATLAB software is used for detection and counting of TB bacilli. Hue color component based approach is used for segmentation of bacilli by thresholding hue range. Shape characterization is used to declare bacilli valid or not. Other artifacts are removed by thresholding the area and perimeter and shape characterizations. Clumps of bacilli are detected using area, perimeter and shape characterizations. Counting is done after segmentation of bacilli and clumps of bacilli. In this work results presented for several images taken from different patients shows that the proposed Scheme detects the presence of TB accurately. |
---|---|
DOI: | 10.1109/ICOMICON.2017.8279128 |