Differentiation of allergenic fungal spores by image analysis, with application to aerobiological counts

The ability of an image analysis routine to differentiate between spores of eleven allergenic fungal genera was tested using analysis based on seven basic and up to 17 more complex features, extracted from digitised images. Fungal spores of Alternaria, Cladosporium, Fusarium, Aspergillus, Penicillium, Botrytis, Epicoccum, Exserohilum, Ustilago, Coprinus and Psilocybe were examined in a series of experiments designed to differentiate between spores at the genus and species level. Linear and Quadratic Discriminant Analysis of feature measurements, recorded for 100 to 1600 spores per taxon, differentiated between genera and species with a high level of accuracy. Genus comparisons using only seven basic features resulted in 98% accuracy for the recognition of conidia belonging to Cladosporium, Fusarium and Epicoccum. Differentiation between conidia of Aspergillus and Penicillium was the least reliable, with 56% of Aspergillus conidia correctly identified and 41% misidentified as Penicillium. At the species level, conidia of Cladosporium macrocarpum, Fusarium moniliforme (microconidia), F. oxysporum (microconidia), F. solani (macroconidia), Alternaria helianthi and A. brassicae were consistently identified with 86--100% accuracy. Reduced levels of accuracy in the identification of spores by image analysis reflected similarities between species in their spore morphology. The application of image analysis to aerobiological counting methods is discussed in relation to the results obtained.