Automated identification of centromere position and centromere index(CI) of human chromosome images

Abstract

Automated chromosome classification is an essential task in cytogenetics. The genetic disorders and abnormalities that occur to the future generation can be predicted through analysing the various characteristics of the chromosomes. The chromosome classification is mainly based on geometric and morphological features. An effective algorithm for chromosome geometric feature extraction is presented. The geometric features of the chromosome are length and centromere. The morphological features are banding pattern. The paper deals with chromosome length and centromere position. The centromere plays an important role to determine the position of P Arm and Q Arm. The P Arm and Q Arm are calculated. The total length is calculated by the sum of P Arm and Q Arm. The proposed algorithm helps in calculating length by curve fitting method which is based on the skeletonization algorithm. The centromere position is identified by finding the concave and convex points on chromosome images.