cDNA microarray image processing using fuzzy vector filtering framework

  • Authors:

  • Rastislav Lukac;Konstantinos N. Plataniotis;Bogdan Smolka;Anastasios N. Venetsanopoulos

  • Affiliations:

  • The Edward S. Rogers Sr. Department of Electrical and Computer Engineering, University of Toronto, 10 King's College Road, Toronto ON, Canada M5S 3G4;The Edward S. Rogers Sr. Department of Electrical and Computer Engineering, University of Toronto, 10 King's College Road, Toronto ON, Canada M5S 3G4;Department of Automatic Control, Silesian University of Technology, Akademicka 16 Str., 44-101 Gliwice, Poland;The Edward S. Rogers Sr. Department of Electrical and Computer Engineering, University of Toronto, 10 King's College Road, Toronto ON, Canada M5S 3G4

  • Venue:
  • Fuzzy Sets and Systems
  • Year:
  • 2005

Quantified Score

Hi-index 0.20

Visualization

Abstract

This paper presents a novel filtering framework capable of processing cDNA microarray images. The proposed two-component adaptive vector filters integrate well-known concepts from the areas of fuzzy set theory, nonlinear filtering, multidimensional scaling and robust order-statistics. By appropriately setting the weighting coefficients in a generalized framework, the method is capable of removing noise impairments while preserving structural information in cDNA microarray images. Noise removal is performed by tuning a membership function which utilizes distance criteria applied to cDNA vectorial inputs at each image location. The classical vector representation, adopted here for a two-channel processing task, as well as a new color-ratio model representation are used. Simulation studies reported in this paper indicate that the proposed adaptive fuzzy vector filters are computationally attractive, yield excellent performance and are able to preserve structural information while efficiently suppressing noise in cDNA microarray data.