New machine-learning-based techniques for DNA microarray image segmentation.
Date of Award
CC BY-NC-ND 4.0
Microarray technology, which provides detailed and abundant information about biological experiments, is a significant achievement in the history of biology. One of the key issues in the microarray processing is to extract quantitative information from the spots, which represent the genes in the experiments. The process of identifying the spots and separating the foreground from the background is known as microarray image segmentation. In this thesis, we present two methods for microarray image segmentation. First, we conduct an in-depth analysis of the influence of important factors on clustering-based microarray image segmentation algorithms. Based on our analysis, we present an optimized clustering-based algorithm for microarray image segmentation, which exploits more than one feature to gain better results comparing to the state-of-the-art clustering-based algorithms. We also consider the fact that most of the spots in a microarray image are ellipses in shape, and hence introduce a novel adaptive ellipse method. This method shows various advantages when compared to the adaptive circle method, one of the most used approaches in microarray image segmentation. The simulations on real-life microarray images show that our method is capable of extracting information from the images which is ignored by the traditional adaptive circle method, and hence showing more flexibility. Paper copy at Leddy Library: Theses & Major Papers - Basement, West Bldg. / Call Number: Thesis2004 .Q26. Source: Masters Abstracts International, Volume: 43-03, page: 0887. Adviser: Luis Rueda. Thesis (M.Sc.)--University of Windsor (Canada), 2004.
Qin, Li., "New machine-learning-based techniques for DNA microarray image segmentation." (2004). Electronic Theses and Dissertations. 2842.