The segmentation of meristematic Allium cell images and extraction of nuclei features for the purpose of mitotic index evaluation

Author

Goclawski, Jaroslaw ; Sekulska-Nalewajko, Joanna ; Aniol, Patryk

Author_Institution

Comput. Eng. Dept., Tech. Univ. of Lodz, Lodz

fYear

2008

fDate

21-24 May 2008

Firstpage

123

Lastpage

128

Abstract

This paper presents the segmentation method of Allium cell images preceding the extraction of nuclei features that are capable to distinguish interphase and mitotic events. The segmentation consists in the isolation of cell regions and then local edge-based thresholding of nuclei parts including the corrections of observed artefacts. The ability of the separation of mitotic phases groups was tested for the nuclei textural and shape features using information gain ratio and Bayesian classification error.

Keywords

biology computing; botany; cellular biophysics; feature extraction; image segmentation; Bayesian classification error; cell region isolation; information gain ratio; interphase events; local edge-based thresholding; meristematic Allium cell image segmentation; mitotic events; mitotic index evaluation; nuclei feature extraction; Cameras; Charge coupled devices; Charge-coupled image sensors; Feature extraction; Image segmentation; Optical distortion; Optical filters; Optical microscopy; Shape; Testing; Allium test; Canny filtering; Hough transform; feature extraction; image segmentation; information gain ratio; mitotic index;

fLanguage

English

Publisher

ieee

Conference_Titel

Perspective Technologies and Methods in MEMS Design, 2008. MEMSTECH 2008. International Conference on

Conference_Location

Polyana

Print_ISBN

978-966-2191-00-4

Type

conf

DOI

10.1109/MEMSTECH.2008.4558758

Filename

4558758