Autoradiographic grain counting by fully automated image analysis.


A method for automated quantitative analysis of single label microautoradiographs applied to DNA repair synthesis is proposed. It uses image analysis of the microscopic field by a Leitz TAS and allows an automated selection of the nuclei to be analysed. The use of a grey level gradient algorithm provides a relatively large independence from external variables such as staining intensity, nuclei density in the field, stability of illumination etc. Random choice of the microscopic field by the computer directed stage displacement and automatic focusing ensures a high degree of objectivity and independence from human intervention. The coefficient of correlation between the automatic and the visual grain count is 0.97 in the range of visually countable grain densities, while the precision of measurement is about 10%. The speed of measurement is satisfactory (40 to 60 s for the analyse of a microscopic field) taking in account that only occasional supervision is needed.


Citations per Year

1,455 Citations

Semantic Scholar estimates that this publication has 1,455 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@article{Kempf1982AutoradiographicGC, title={Autoradiographic grain counting by fully automated image analysis.}, author={Julie Kempf and Claudia Kempf}, journal={Microscopica acta}, year={1982}, volume={86 3}, pages={215-23} }