Chromosome Overlap Segmentation Algorithm Based on Chromosome Outline and Skeleton
- 21 April 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 182-188
Abstract
Karyotype analysis is an effective tool in the diagnosis of chromosomal diseases. Image processing technology enables the analysis and medical description of the number and morphological characteristics of chromosomes. Chromosome images are characterized by the high cohesion, overlapping and nesting, which represents a difficult problem in chromosome image segmentation at this stage. To effectively segment overlapping chromosomes, this paper proposes an improved overlapping chromosome segmentation algorithm based on contour and skeleton. Based on the preprocessed chromosome karyotype map, the algorithm applies the Canny operator for chromosome contour detection and edge information extraction to accurately locate the skeleton and complete the segmentation. We evaluated our algorithm on 200 chromosome images, and the results showed the segmentation accuracy of “+” and “T” shaped overlapping chromosomes to be 97.8% and 96.1%, respectively, while the accuracy of overlapping chromosome segmentation in the whole image was 98.5%. This shows the effectiveness of the algorithm.Keywords
Funding Information
- National Natural Science Foundation of China
This publication has 16 references indexed in Scilit:
- A geometric approach to fully automatic chromosome segmentation2014
- A hybrid technique for overlapped chromosome segmentation of G-band mataspread images automatic2014
- Separation of overlapped chromosomes and pairing of similar chromosomes for karyotyping analysisMeasurement, 2014
- A novel approach for efficient extrication of overlapping chromosomes in automated karyotypingMedical & Biological Engineering & Computing, 2013
- Automatic segmentation of metaphase cells based on global context and variant analysisPattern Recognition, 2008
- Noise Estimation from a Single Image2006
- Bilateral filtering for gray and color images2002
- A classification-driven partially occluded object segmentation (CPOOS) method with application to chromosome analysisIEEE Transactions on Signal Processing, 1998
- A fast parallel algorithm for thinning digital patternsCommunications of the ACM, 1984
- Algorithm AS 136: A K-Means Clustering AlgorithmJournal of the Royal Statistical Society Series C: Applied Statistics, 1979