Development and Validation of a Near-Infrared Spectroscopy Method for the Prediction of Acrylamide Content in French-Fried Potato

This study investigated the ability of near-infrared spectroscopy (NIRS) to predict acrylamide content in French fried potato. Potato flour spiked with acrylamide (50-8000 mu g/kg) was used to determine if acrylamide could be accurately predicted in a potato matrix. French fries produced with various pretreatments and cook times (n = 84) and obtained from quick service restaurants (n = 64) were used for model development and validation. Acrylamide was quantified using gas chromatography-mass spectrometry, and reflectance spectra (400-2500 nm) of each freeze-dried sample Were captured on a Foss XDS Rapid Content Analyzer-NIR spectrometer. Partial least-squares (PLS) discriminant analysis and PLS regression modeling demonstrated that NIRS could accurately detect acrylamide content as low as 50 mu g/kg in the model potato matrix. Prediction errors of 135 mu g/kg (R-2 = 0.98) and 255 mu g/kg (R-2 = 0.93) were achieved with the best PLS models for acrylamide prediction in Russet Norkotah French-fried potato and multiple samples of unknown varieties, respectively. The findings indicate that NIRS can be used as a screening tool in potato breeding and potato processing research to reduce acrylamide in the food supply.