Pixel-by-pixel classification of MFISH images

Multiplex Fluorescence In-Situ Hybridization (M-FISH) is a recently developed chromosome imaging method in which each chromosome is labelled with 5 fluors (dyes) and is also counterstained with DAPI. This paper proposes an automatic pixel by pixel classification algorithm for M-FISH images using a Bayes classifier. The M-FISH pixel classification was approached as a 25 class 6 feature pattern recognition problem. The classifier was trained and tested on non-overlapping data sets and an overall classification accuracy of 95% was obtained.

[1]  K. Castleman,et al.  FISH image analysis , 1996 .

[2]  David G. Stork,et al.  Pattern Classification , 1973 .

[3]  D. Ward,et al.  Karyotyping human chromosomes by combinatorial multi-fluor FISH , 1996, Nature Genetics.

[4]  Shigeo Abe DrEng.  Pattern Classification , 2001, Springer London.

[5]  Alan C. Bovik,et al.  Minimum entropy segmentation applied to multi-spectral chromosome images , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[6]  M. L. Beau,et al.  One FISH, two FISH, red FISH, blue FISH , 1996, Nature Genetics.