Effective quantification of gene expression levels in microarray images using a spot-adaptive compound clustering-enhancement-segmentation scheme

  • Authors:

  • Antonis Daskalakis;Dionisis Cavouras;Panagiotis Bougioukos;Spiros Kostopoulos;Pantelis Georgiadis;Ioannis Kalatzis;George Kagadis;George Nikiforidis

  • Affiliations:

  • Department of Medical Physics, School of Medicine, University of Patras, Rio, Greece;Medical Signal and Image Processing Lab, Department of Medical Instruments Technology, Technological Educational Institute of Athens, Athens, Greece;Department of Medical Physics, School of Medicine, University of Patras, Rio, Greece;Department of Medical Physics, School of Medicine, University of Patras, Rio, Greece;Department of Medical Physics, School of Medicine, University of Patras, Rio, Greece;Medical Signal and Image Processing Lab, Department of Medical Instruments Technology, Technological Educational Institute of Athens, Athens, Greece;Department of Medical Physics, School of Medicine, University of Patras, Rio, Greece;Department of Medical Physics, School of Medicine, University of Patras, Rio, Greece

  • Venue:
  • ICCSA'07 Proceedings of the 2007 international conference on Computational science and its applications - Volume Part III
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

A spot-adaptive compound clustering-enhancement-segmentation (CES) scheme was developed for the quantification of gene expression levels in microarray images. The CES-scheme employed 1/griding, for locating spot-regions, 2/Fuzzy C-means clustering, for segmenting spots from background, 3/ background noise estimation and spot's center localization, 4/emphasizing of spot's outline by the CLAHE image enhancement technique, 5/segmentation by the SRG algorithm, using information from step 3, and 6/microarray spot intensity extraction. Extracted intensities by the CES-Scheme were compared against those obtained by the MAGIC TOOL's SRG. Kullback-Liebler metric's values for the CES-Scheme were on average double than MAGIC TOOL's, with differences ranging from 1.45bits to 2.77bits in 7 cDNA images. Coefficient-of-Variation results showed significantly higher reproducibility (p