Related Experiment Videos
Feature analysis and centromere segmentation of human chromosome images using an iterative fuzzy algorithm
Parvin Mousavi1, Rabab Kreidieh Ward, Sidney S Fels
1Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, Canada.
IEEE Transactions on Bio-Medical Engineering
|April 11, 2002
Summary
This study presents a new fuzzy algorithm for segmenting centromeres in human chromosomes. This method accurately classifies homologous chromosomes using centromere intensity and morphology.
Area of Science:
- Genetics
- Biotechnology
- Image Analysis
Background:
- Accurate classification of homologous chromosomes is crucial for cancer genetics research.
- Centromere intensity is a key feature for differentiating homologous chromosomes.
- Effective segmentation of centromeres is a prerequisite for homolog classification.
Purpose of the Study:
- To develop and validate an iterative fuzzy algorithm for segmenting centromeres in human chromosome images.
- To utilize segmented centromere intensities and morphological differences for homologous chromosome classification.
Main Methods:
- An iterative fuzzy algorithm was developed, assigning fuzzy membership values to pixels.
- The algorithm iteratively updates and minimizes an error function for centromere segmentation.
- Fluorescence in-situ hybridization (FISH) technique was used for chromosome preparation.
Main Results:
- The developed algorithm successfully segmented centromeres from human chromosome images.
- Chromosome 22, a highly heteromorphic chromosome, was used for verification.
- Classification of chromosome 22 homologs based on centromere intensity and morphology showed complete agreement.
Conclusions:
- The iterative fuzzy algorithm provides accurate centromere segmentation.
- The method effectively classifies homologous chromosomes using centromere features.
- This approach validates the algorithm's utility in advanced genetic studies.