Manually counting cell colonies, especially those that originate from fibro
blast cell lines, is a time-consuming, eye-straining and tedious task in wh
ich consistency of counting is difficult to maintain. In this paper we pres
ent a novel model-based image segmentation method, which employs prior know
ledge about the shape of a colony with the aim to automatically detect isol
ated, touching and overlapping cell colonies of various sizes and intensiti
es. First, a set of hypothetical model instances is generated by using a ro
bust statistical approach to estimate the model parameters and a novel conf
idence measure to quantify the difference between a model instance and the
underlying image. Second, the model instances matching the individual colon
ies in the image are selected from the set by a minimum description length
principle. The procedure was applied to images of Chinese hamster lung fibr
oblast cell line DC3F, which forms poorly defined or 'fuzzy' colonies. The
correlation with manual counting was determined and the cell survival curve
s obtained by automated and manual counting were compared. The results obta
ined show that the proposed automatic procedure was capable to correctly id
entify 91% of cell colonies typical of mammalian cell lines.