Skip to main content
ARS Home » Southeast Area » New Orleans, Louisiana » Southern Regional Research Center » Food and Feed Safety Research » Research » Publications at this Location » Publication #381391
Research Project: Use of Classical and Molecular Technologies for Developing Aflatoxin Resistance in Crops

Location: Food and Feed Safety Research

Title: Identification of aflatoxin contamination in corn kernels based on near infrared hyperspectral imaging

Author
item TAO, FEIFEI - Mississippi State University
item YAO, HAIBO - Mississippi State University
item HRUSKA, ZUZANA - Mississippi State University
item KINKAID, RUSSELL - Mississippi State University
item Rajasekaran, Kanniah - Rajah

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 12/18/2020
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Aflatoxin contamination in agricultural and food products is a global food safety concern. Near infrared hyperspectral imaging over the 900-2500 nm spectral range was investigated for its potential to identify post-harvest aflatoxin contamination in corn kernels, in a rapid and non-destructive manner. A total of 600 kernels were included with 3 treatments, namely, 200 kernels inoculated with Aspergillus flavus AF13 (aflatoxigenic) strain, 200 kernels inoculated with Aspergillus flavus AF36 (non-aflatoxigenic) strain and 200 kernels inoculated with sterile distilled water as control. One hundred kernels from each treatment were incubated for 5 and 8 days, to obtain samples with various levels of contamination, confirmed with toxin analysis. The mean spectrum of each kernel was extracted from the endosperm side and used for developing the partial least-squares discriminant analysis (PLS-DA) model. Based on the classification threshold of 20 ppb, the established full spectral PLS-DA model achieved prediction accuracies of 92.1% and 77.8% for the aflatoxin-negative class and -positive class, respectively. The corresponding overall prediction accuracy was 88.7%. Simplification of full spectral model was also conducted by selecting the most informative spectral features. The results indicated the usefulness of near infrared hyperspectral imaging in identifying aflatoxin-contaminated corn kernels and the possibility of developing a low-cost multispectral screening system. Visualization of aflatoxin contamination map is also presented in this study.