A Novel Method for Centromere and Length Detection in Microscopic Images of Human Chromosomes

Many genetic disorders or abnormalities that may occur in the future generations can be predicted throughanalyzing the shape and morphological characteristics of the human chromosomes. This is usually carried out by an expert, inspecting the Karyotype of the patients. A Karyotype is a particular table that presents the chromosome images in a standard format. To generate a Karyotype, it is necessary to identify each of the 23 pairs of the chromosomes within the microscopic images first. The main step to automate this procedure is the definition of some morphological features for each chromosome. The most common features used for chromosomes identification includes the location of the Centromere and the length of the chromosome. Many other important features, such as Centromeric Index, are usually extracted from the Centromere and length. In this paper, a novel and effective algorithm for Centromere locating and length calculation for the human chromosomes is presented. The proposed algorithm uses the fact that the centromere is the narrowest part of the chromosome. By defining a gray level mask (GLM), which is a linearly varying gray level image along the chromosome longitudinal directionand multiplying it to the binary version of the chromosome image, it is shown that the global minimum in the histogram of the resulted image indicates the location of the centromere. The data set used in this work was provided by the Tesi-Imaging srl in Milan, Italy. A mean value of the absolute error of 3.6 and 5.2 pixels was obtained in identification of the chromosome centromere and length respectively by the proposed method.