A novel use of a geometric morphometric technique to distinguish human parasite eggs of twelve different species

Abstract

Copro-microscopic diagnostic methods are the most common approach for screening patients with parasitic infections. However, expertise is required to identify helminthic eggs from fecal specimens. Consequently, new methods are required to support accurate species identification. Novel technologies have recently been developed for the classification of organisms, including geometric morphometric (GM) approaches. In this study, the outline-based GM approach was used to distinguish the eggs of 12 common human parasite species, including Ascaris lumbricoides, Trichuris trichiura, Enterobius vermicularis, hookworm, Capillaria philippinensis, Opisthorchis spp., Fasciola spp., Paragonimus spp., Schistosoma mekongi, Taenia spp., Hymenolepis diminuta and Hymenolepis nana. The GM analysis revealed that the size cannot be used as the main variable in the identification of parasite species at the egg stage, producing only 30.18% overall accuracy. However, comparisons of shape based on the Mahalanobis distances between pairs of parasite species showed significant differences in all pairs (p < 0.05). The shape analysis produced 84.29% overall accuracy. This is the first time that outline-based GM has been preliminarily confirmed as a valuable approach to support copro-microscopic analysis, in order to effectively screen helminth eggs. However, further studies with a larger set of helminth eggs and artefacts should be carried out to increase confidence in the identification of parasite species in the absence of local experts.