Interactive classification of porcine Eimeria spp. by computer-assisted image analysis

Digitalized pictures of oocysts of Eimeria scabra, E. polita, E. debliecki, E. suis, and E. porci were used to develop a program routine that allows semi-automatic interactive classification of porcine Eimeria spp. with a computer-assisted image analysis system. Morphological parameters (shape, color, length and width of oocysts and length and width of sporocysts) were assayed, transformed to color and shape factors and used to calculate classification indices (G1, SP). Thresholds were defined for these indices to allocate oocysts to groups of species and successively to single species. In more than 97% of cases the program routine allocated digitalized oocysts (n = 175) to the respective species. E. suis was always correctly classified. Interactive classification proved to be convenient, fast, precise and largely free of individual bias.