Integration of handheld NIR and machine learning to "Measure & Monitor" chicken meat authenticity

Hadi Parastar, Geert van Kollenburg, Yannick Weesepoel, Andre van den Doel, Lutgarde Buydens, Jeroen Jansen
2020 Food Control  
1 By combining portable, handheld near-infrared (NIR) spectroscopy with state-of-the-2 art classification algorithms, we developed a powerful method to test chicken meat 3 authenticity. The research presented shows that it is both possible to discriminate fresh from 4 thawed meat, based on NIR spectra, as well as to correctly classify chicken fillets according to 5 the growth conditions of the chickens with good accuracy. In all cases, the random subspace 6 discriminant ensemble (RSDE) method
more » ... gnificantly outperformed other common 7 classification methods such as partial least squares-discriminant analysis (PLS-DA), artificial 8 neural network (ANN) and support vector machine (SVM) with classification accuracy of 9 >95%. This study shows that handheld NIR coupled with machine learning algorithms is a 10 useful, fast, non-destructive tool to identify the authenticity of chicken meat. By comparing 11 and combining different protocols to measure the NIR spectra (i.e., through packaging and 12 directly on meat), we show the possibilities for both consumers and food inspection 13 authorities to check the authenticity and origin of packaged chicken fillet. 14
doi:10.1016/j.foodcont.2020.107149 fatcat:2t5jvf6nbvf5neoynz3qhaifyi