Available Datasets:

Chromosome Images for Segmentation

Chromosome Images for Classification

Automatic Karyotyping:

Segmentation

Classification


Automatic Karyotyping: Segmentation

E. Grisan, E. Poletti, A. Ruggeri. Automatic segmentation and disentangling of chromosome in Q-band prometaphase images. IEEE Trans Inf Technol Biomed 13(4):575-81, Jul 2009.
E. Grisan, E. Poletti, A. Ruggeri. An Improved Segmentation of Chromosomes in Q-Band Prometaphase Images Using a Region Based Level Set. WC 2009, IFMBE Proceedings 25/XI, pp. 748–51, Springer-Verlag, Berlin Heidelberg 2009. 

 

Chromosome karyotyping analysis is an important screening and diagnostic procedure routinely performed in clinical and cancer cytogenetic labs. Chromosome are first stained with a fluorescent dye, and then imaged through a microscope for subsequent analysis and classification. Each chromosome in the image has to be identified and assigned to one of 24 classes: the result is the so-called karyotype image in which all chromosomes in a cell are graphically arranged according to an international system for cytogenetic nomenclature (ISCN) classification.

      

The figure shows a typical PAL resolution (768 x 576, 8 bits/pixel) Qbanding prometaphase image and the corresponding karyotype. Individual chromosomes only appear as distinct bodies towards the end of the cell division cycle, at prophase, when they are long string-like objects, contracting and separating at metaphase, just before cell division. The intermediate stage of contraction between prophase and metaphase is called prometaphase.
Most of the studies aimed at the development of cytogenetics systems for analysis of banded chromosome preparations, have concentrated on metaphase chromosomes, avoiding the segmentation difficulties arising from touches and overlaps in the prophase and prometaphase cells.

We proposed an automatic procedure to obtain the separated chromosomes, which are then ready for a subsequent classification step. The segmentation is carried out by means of a space variant thresholding scheme, which proved to be successful even in presence of hyper- or hypo-fluorescent regions in the image. Then a greedy approach is used to identify and resolve touching and overlapping chromosomes, based on geometric evidence and image information.

We have developed an algorithm able to automatically identify chromosomes in metaphase images, taking care of a first segmentation step and then of the disentanglement of chromosome clusters by resolving separately adjacencies and overlaps with a greedy approach, that ensures that at each step only the best split of a blob is performed.

The performance of our methods are better or comparable to the best of other methods reported in the literature. To the best of our knowledge, this the first methods to simultaneously tackle segmentation, overlaps and adjacencies with performance on each task higher than 90%, hence providing a tool able to automatically analyze an image, and whose results can be handed over wit minimal human intervention to a classifier for automatic karyotyping.


return to top