Teff is a gluten-free cereal with valuable nutritional composition, making it a potential target for fraud. Therefore, the food industry and consumers require rapid fraud detection methods like spectroscopy. Thus, the performance of NIR in detecting and quantifying teff flours adulterated with other cereals was evaluated
the NIR spectra were also used to predict the proximal composition. Adulteration with rice, whole wheat, rye, and oats followed a simplex-lattice mixture design {5,4}, where five corresponds to the number of mixture components and four different proportions are assigned for each component. The limits were 65 %-100 % for teff and 0-35 % for adulterants. The proximate composition of pure flours was determined using reference methods, and the composition of mixtures was determined by mass balance. Partial Least Squares (PLS) regression models of the near-infrared spectra were constructed with low limits of quantification (11 %, 0.68 %, 0.20 %, 0.08 %, and 0.36 % for teff content, carbohydrates, total lipids, ash, and crude protein, respectively). Therefore, NIR has proven to be a valuable tool for quality control and fraud detection in teff flours, offering enhanced efficiency.