A Pattern Classification Approach to DNA Microarray Image Segmentation

  • Authors:
  • Luis Rueda;Juan Carlos Rojas

  • Affiliations:
  • School of Computer Science, University of Windsor, Windsor, Canada N9B 3P4;Department of Computer Science, University of Concepción, Concepción, Chile

  • Venue:
  • PRIB '09 Proceedings of the 4th IAPR International Conference on Pattern Recognition in Bioinformatics
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

A new method for DNA microarray image segmentation based on pattern recognition techniques is introduced. The method performs an unsupervised classification of pixels using a clustering algorithm, and a subsequent supervised classification of the resulting regions. Additional fine tuning includes detecting region edges and merging, and morphological operators to eliminate noise from the spots. The results obtained on various microarray images show that the proposed technique is quite promising for segmentation of DNA microarray images, obtaining a very high accuracy on background and noise separation.