Technical reports

Gelfish – graphical environment for labelling FISH images

Boaz Lerner, Seema Dhanjal, Maj Hultén

May 1999, 20 pages

DOI: 10.48456/tr-465

Abstract

Dot counting in flourescence in-situ hybridization (FISH) images that relies on an automatic focusing method for obtaining clearly defined images is prone to errors. Our recently developed system has dispensed with automatic focusing, and insteaqd relies on a larger statistical sample of the specimen at a fixed focal plane. The system is based on well-discriminating features to represent the signals and a neural network classifier to discriminate between artifacts and valid signal data. Results showed that nearly 90% of valid signals and artifacts of two flourophores within 400 FISH images were correctly classified. To train the classifier, accurate labelling of the image is required. GELFISH is a Graphical Environment for Labelling FISH images that enables the labelling of FISH signals and the rejection of unanalysable nuclei simply and rapidly. Feedback provided by the environment allows the user to correct the results of labelling effortlessly by clicking GELFISH buttons using the mouse. Furthermore, GELFISH is flexible and can be modified easily for additional FISH applications. Implemented using popular software, the environment can be employed on any computer by any user.

Full text

PDF (1.3 MB)

BibTeX record

@TechReport{UCAM-CL-TR-465,
  author =	 {Lerner, Boaz and Dhanjal, Seema and Hult{\'e}n, Maj},
  title = 	 {{Gelfish -- graphical environment for labelling FISH images}},
  year = 	 1999,
  month = 	 may,
  url = 	 {https://www.cl.cam.ac.uk/techreports/UCAM-CL-TR-465.pdf},
  institution =  {University of Cambridge, Computer Laboratory},
  doi = 	 {10.48456/tr-465},
  number = 	 {UCAM-CL-TR-465}
}