New method uses near-infrared hyperspectral imaging to visualize the mycotoxin. Fumonisins, produced by fungi of the genus Fusarium, contaminate grains in the field and are not destroyed by thermal processing. Hence they are among the worst threats to corn production in Brazil. The NIR-HSI technique also makes it possible to identify contaminated batches and prevent cross-infection during corn storage. Scientists from Embrapa and the Federal University of Minas Gerais (UFMG) developed an innovative method to detect the presence of fumonisin in corn kernels without the need for grinding or chemical reagents, which reduces costs and makes the process environmentally healthier. The technique uses near-infrared hyperspectral images (NIR-HSI), which combines chemistry and precision agriculture precepts, to identify and quantify such mycotoxin, considered one of the biggest obstacles to corn production in Brazil as it contaminates the kernels while the grains are still in the field and is not destroyed by thermal processing. Fumonisins are mostly produced by fungi from the genus Fusarium and, as they have wide distribution, broad occurrence and high toxicity, they are considered the worst mycotoxins among those produced by such microorganisms. Once associated with the mathematical model for multivariate image analysis, NIR-HSI allows observers to quickly identify and quantify fumonisins (which are invisible to the naked eye) directly in corn kernels and without destroying corn samples. “NIR-HSI technology works based on the principle of diffuse reflectance, which depends on the chemical and structural properties of the material. It is a non-destructive approach to obtain spatially distributed spectra, which allows you to visualize and locate chemical changes in any complex system pixel by pixel”, Maria Lúcia Simeone, a researcher at Embrapa Maize and Sorghum, explains. Innovation in mycotoxin detection The method currently used to quantify fumonisins is expensive, complex, time-consuming, and requires a high level of technical knowledge and that samples are ground. On top of such disadvantages there is the fact that the chemical reagents used to perform the analysis are toxic, which results in harm to the health of the analyst and of the environment. According to Simeone, the new method is much faster, does not use chemicals, does not destroy the sample and has a lower cost. “It works through an algorithm that was built from spectral and spatial information obtained from NIR-HSI equipment using different corn samples, as the data depends on the interaction between electromagnetic radiation and atoms or molecules in the sample under analysis,” she reports. The researcher also highlights that the results obtained with the NIR-HSI technique were surprising, especially because they allowed the identification of contaminated batches and prevention of cross-infection in storage. “This methodology has the potential to transform the way we quantify and control fumonisin, ensuring food quality and safety,” she adds. Benefits for the production chain The new technique brings several benefits to the whole corn production chain: Speed: The quantification of the fumonisin content is quickly performed in just 30 seconds, which allows a higher number of samples to be analyzed in a shorter time, and a prompter response in case of contamination. Cost reduction: The technique is more economical than traditional methods, as it dispenses with grinding or use of chemical reagents. Non-destruction: The analysis does not damage the sample, allowing the analysis to be perfomed directly on the grains, which can be used later. A safer future for corn consumption The study published in the journal Brazilian Journal of Biology represents a significant advance in the area of food security. “By allowing the quick and direct detection of the fumonisin content in corn kernels, this new methodology contributes to ensuring food quality and safety, thus protecting the health of consumers and of animals”, observes Renata Pereira da Conceição, a postgraduate student at UFMG. Photo by: Renata Silva For the Embrapa researcher Valéria Aparecida Vieira Queiroz, “with this technology, it is possible to develop more efficient strategies to control fumonisins in corn, reducing production losses, enabling the segregation of batches of samples, and ensuring safer food for the population”. Everaldo Medeiros, a researcher at Embrapa Cotton, explains that the technique generates a sort of “chemical image of the object” by combining chemometric data processing techniques. This allows the users to explore innovative applications for agriculture based on green chemistry and precision agriculture concepts, which place Embrapa and partners at the frontier of innovation with applying NIR-HSI images. The researcher Everaldo Medeiros talks about the technique “Our participation was to study the best image configurations for the measurements of fumonisins directly in the corn seeds. The results allowed the automatic detection and quantification of mycotoxins with higher sensitivity and speed than the techniques currently in use offer”, Medeiros concludes. Paper and authors The paper entitled Determination of fumonisin content in maize using near-infrared hyperspectral imaging (NIR-HSI) technology and chemometric methods was authored by: R. R. P. Conceição (Federal University of Minas Gerais, Institute of Biological Sciences, Belo Horizonte, MG, Brazil); V. A. V. Queiroz (Embrapa Maize and Sorghum, Sete Lagoas, MG, Brazil); E. P. Medeiros (Embrapa Cotton, Campina Grande, PB, Brazil); J. B. Araújo (Embrapa Cotton, Campina Grande, PB, Brazil); D. D. S. Araújo (Embrapa Maize and Sorghum, Sete Lagoas, MG, Brazil); R. A. Miguel (Embrapa Maize and Sorghum, Sete Lagoas, MG, Brazil); M. A. R. Stoianoff (Federal University of Minas Gerais, Institute of Biological Sciences, Belo Horizonte, MG, Brazil) and M. L. F. Simeone (Embrapa Maize and Sorghum, Sete Lagoas, MG, Brazil).