Detección automatizada de adulteraciones en zumos de frutas mediante HS-GC-IMS

Abstract

Fruit juices are one of the most widely consumed beverages worldwide and their production is subject to European regula tions. However, this food is often adulterated by adding water, sugars or other less expensive fruit juices. The latter is very com mon dueto the greater difficulty to be detected. Therefore, the present study aimed to develop a methodology based on Head space-Headspace-Gas Chromatography-Ion Mobility Spectrom etry (Hs-GC-IMS) in combination with machine learning (ML) algorithrns to reliably detect fruit juice adulterations. For this purpose, three types of 100% squeezed fruit juices (pineapple, orange and apple) were evaluated and adulterated with grape juice in different ratios (5%, 10%,15%,20%,30%,40% and 50%). The exploratory analysis revealed a clustering trend to classifies the samples according to the type of juice analyzed. The super vised analysis, based on the development of machine leaming models for the detection of adulteration, achieved successful performance, obtaining the best result for the support vector machines with an accuracy of 95.45% in the test set. For adulter ant quantification the best results were obtained with support vector regression showing an R2 of 0.987 anda RMSE of 1.831 for the test set. In addition, a simple web application has been developed where the trained models can be used, so any re searcher can apply the method to detect this type of adulteration.