Table of Contents Author Guidelines Submit a Manuscript
Journal of Chemistry
Volume 2015, Article ID 485864, 10 pages
Research Article

Assessing Risk of Fumonisin Contamination in Maize Using Near-Infrared Spectroscopy

1Université de Toulouse, Ecole d’Ingénieurs de Purpan, INPT, LCA, 31076 Toulouse Cedex 03, France
2INRA, UMR1010 CAI, 31030 Toulouse, France
3INRA, UMR1331, Research Centre in Food Toxicology (Toxalim), 31027 Toulouse, France
4Université de Toulouse, ENVT, INP, Toxalim, 31076 Toulouse, France
5Département Sciences Agronomiques et Agroalimentaires, Université de Toulouse, Ecole d’Ingénieurs de Purpan, INPT, 31076 Toulouse Cedex 03, France

Received 6 January 2015; Revised 23 March 2015; Accepted 24 March 2015

Academic Editor: Leiqing Pan

Copyright © 2015 Cecile Levasseur-Garcia et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


Fumonisins are major mycotoxins found worldwide in maize and maize products. Because of their toxicity for both human and animals, European Union regulations were created to fix the maximal fumonisin B1 and B2 content allowed in foods and feeds. Unfortunately, directly measuring these mycotoxins by current analytical techniques is tedious and expensive and most measurement methods do not lend themselves to online control. Alternative approaches to chemical analysis have been developed and involve models that allow the mycotoxin contamination to be predicted based on environmental conditions and analysis by near-infrared (NIR) spectroscopy. In the present work, we use NIR spectroscopy to determine the fumonisin and fungal contents of 117 samples of maize. The determination coefficient between fumonisin and fungal-biomass content was 0.44. We establish herein a threshold for the number of CFUs for fungal biomass beyond which the fumonisin content is likely to exceed the European regulatory level of 4000 μg/kg. In addition, we determine the fungal content by using a NIR-spectroscopy model that allows us to sort samples of maize. Upon calibration, the percentage of well-classified samples was 96%, which compares favorably to the 82% obtained by independent verification.