Multichannel Segmentation of cDNA Microarray Images using the Bayes Classifier

Microarray technology provides a powerful tool for the quantification of the expression level for a large number of genes simultaneously. Image analysis Is a crucial step for data extraction of microarray gene expression experiments. In this paper we propose a supervised method for the segmentation of microarray Images. The Bayes classifier Is employed for a pixel by pixel classification. The proposed method classifies the pixels of the Image In two classes, foreground and background pixels. For this task, an Informative set of features Is used from both green and red channels. The method Is evaluated using a set of 5184 spots (consisting of ~15000000 pixels), from the Stanford microarray database (SMD) and the reported classification accuracy Is 82 %.