EARLY DETECTION OF TOXIGENIC FUNGI ON MAIZE KERNELS BY SPECTRAL IMAGING ANALYSIS.

Maize (Zea mays L.) is an important cereal, used for food and for animal feed. Several toxigenic fungi can infect maize in the field and in the post harvest conditions. In particular, maize is susceptible to infection by fungi belonging to Fusarium and Aspergillus. Several studies have assayed non-destructive, spectral methods to detect fungal contamination and toxins on cereals. The aim of this work is the early detection of toxigenic fungi on whole maize kernels, the discrimination between healthy and diseased kernels and the determination of damage degree by using a Spectral Imaging -based method. A desktop spectral scanner equipped with an imaging based spectrometer ImSpector-Specim V10, working in the visible-near infrared spectral range (400-1000 nm) was used in the esperiments. Four pure toxigenic fungal strains were imaged, A. parasiticus, A. flavus, A. niger and F. graminearum. The results obtained indicate that spectral imaging can discriminate between different fungal strains, producing a specific spectral fingerprint for each one. These fungi have been also inoculated on maize kernels and were imaged from 24 h to 7 days of growth. Changes in reflectance of kernels were observed during fungal growth. Spectral imaging could be able to discriminate maize kernels infected with toxigenic fungi from uninfected controls; this technique could be used for an early detection of toxigenic fungi on cereals