Computer image analysis of combinatorial multi-fluor FISH

Abstract

Epifluorescence filter sets and computer software for the detection and discrimination of up to 27 different DNA probes using multiplex-fluorescence in situ hybridization (M-FISH) have been developed recently. This paper focuses on a more detailed description of the specific software developed for the analysis of M-FISH experiments. Crucial steps in the evaluation process such as standardization of image acquisition, chromosome segmentation, fluorescence background estimation, accurate measurement of fluorescence intensities and definition of thresholds will be described. The application of M-FISH and comparative genomic hybridization (CGH) for the detection of chromosomal abnormalities in squamous cell cancer of the head and neck is also reported. Finally, we try to define an acceptable ‘standard’ for a fully automated multicolor imaging system.