DocumentCode
2352187
Title
An Unparagoned Application for Red Blood Cell Counting using Marker Controlled Watershed Algorithm for Android Mobile
Author
Karunakar, Yogesh ; Kuwadekar, Alhad
Author_Institution
Univ. of Mumbai, Mumbai, India
fYear
2011
fDate
14-16 Sept. 2011
Firstpage
100
Lastpage
104
Abstract
Many diseases are diagnosed based on cell count done by clinical examination. The methods which are being used in pathological laboratories for counting cells in a sample are the traditional manual and tedious ones. This procedure is laborious and dependent on the judgment and skill of the operator. Moreover the judgment of the operator may vary in relation to fatigue and other physical conditions. Importantly these techniques consume long periods of time which may affect diagnosis. In order to improve the efficiency and veracity of diagnosis, this paper presents an algorithm that automatically counts the cells using windows based applications in mobile phones. The algorithm uses the quintessential preprocessing techniques to remove the graininess in the sample images, enhances the contrast between the cytoplasm and nucleus and extra cellular components [15]. Subsequently, the segmentation algorithm based on Otsu´s threshold, morphological operations, and marker controlled watershed algorithm and cell size considerations is performed. Pathologists can now prepare and image thousands of samples and save on time per day using this automation.
Keywords
cellular biophysics; diseases; image segmentation; medical image processing; mobile handsets; Android mobile; Otsu´s threshold; cytoplasm; diagnosis efficiency; diagnosis veracity; diseases; extracellular components; marker controlled watershed algorithm; mobile phones; nucleus; red blood cell counting; segmentation algorithm; unparagoned application; Educational institutions; Image resolution; Image segmentation; Microscopy; Mobile communication; Shape; Transforms;
fLanguage
English
Publisher
ieee
Conference_Titel
Next Generation Mobile Applications, Services and Technologies (NGMAST), 2011 5th International Conference on
Conference_Location
Cardiff
ISSN
2161-2889
Print_ISBN
978-1-4577-1080-3
Type
conf
DOI
10.1109/NGMAST.2011.27
Filename
6082002
Link To Document