Automated detection of the cut-points for the separation of overlapping chromosomes

Since the birth of the automated karyotyping systems by the aid of computers, building a fully automated chromosome analysis system has been an ultimate goal. Along with many other challenges, automating chromosome classification and segmentation has been a major challenge especially due to overlapping and touching chromosomes. The earlier reported methods for disentangling the chromosome overlaps have limited success as they are sensitive to scale variations, computationally complex, use only color information in case of multispectral imaging and most of them are limited to separation of single overlap formed by two chromosomes in a cluster. This paper proposes first step towards the extrication of the overlapping chromosomes for the karyotyping of the metaphase image by automated detection of the required cut-points. The proposed simple but novel and efficient approach to automatically detect the cut points is based on the computational geometry of the pixels on the boundary of the overlapping cluster. Contribution and novelty of this work is in the ability of the algorithm to successfully identify the cut points in a cluster with multiple chromosomes. System performance was tested and analyzed using a variety of synthesized images from LK1 data base exhibiting various levels of overlapping chromosomes giving an overall accuracy of 100 % in cases of clusters with 1 and 2 overlaps and 88 % in cases of clusters with 3 and 4 overlaps.