Sciweavers

TMI
2008

Feature Normalization via Expectation Maximization and Unsupervised Nonparametric Classification For M-FISH Chromosome Images

13 years 3 months ago
Feature Normalization via Expectation Maximization and Unsupervised Nonparametric Classification For M-FISH Chromosome Images
Multicolor fluorescence in situ hybridization (M-FISH) techniques provide color karyotyping that allows simultaneous analysis of numerical and structural abnormalities of whole human chromosomes. Chromosomes are stained combinatorially in M-FISH. By analyzing the intensity combinations of each pixel, all chromosome pixels in an image are classified. Often, the intensity distributions between different images are found to be considerably different and the difference becomes the source of misclassifications of the pixels. Improved pixel classification accuracy is the most important task to ensure the success of the M-FISH technique. In this paper, we introduce a new feature normalization method for M-FISH images that reduces the difference in the feature distributions among different images using the expectation maximization (EM) algorithm. We also introduce a new unsupervised, nonparametric classification method for M-FISH images. The performance of the classifier is as accurate as the ...
Hyohoon Choi, Alan C. Bovik, Kenneth R. Castleman
Added 15 Dec 2010
Updated 15 Dec 2010
Type Journal
Year 2008
Where TMI
Authors Hyohoon Choi, Alan C. Bovik, Kenneth R. Castleman
Comments (0)