Abstract | ||
---|---|---|
Microarray technology is a powerful tool for analyzing the expression of a large number of genes in parallel. A typical microarray image consists of a few thousands of spots which determine the level of gene expression in the sample. In this paper we propose a method which automatically addresses each spot area in the image. Initially, a preliminary segmentation of the image is produced using a template matching algorithm. Next, grid and spot finding are realized. The position of non-expressed spots is located and finally a Voronoi diagram is employed to fit the grid on the image. Our method has been evaluated in a set of five images consisting of 45960 spots, from the Stanford microarray database and the reported accuracy for spot detection was 93% |
Year | DOI | Venue |
---|---|---|
2006 | 10.1109/IEMBS.2006.260782 | EMBC |
Keywords | DocType | Volume |
gene expression,computational geometry,template matching,image segmentation,genetics,microarray technology,voronoi diagram | Conference | 1 |
ISSN | Citations | PageRank |
1557-170X | 6 | 0.60 |
References | Authors | |
9 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Nikolaos Giannakeas | 1 | 6 | 0.60 |
Dimitrios I Fotiadis | 2 | 49 | 14.82 |
Anastasia S Politou | 3 | 6 | 0.60 |