Rapid Evaluation of Oxidized Fatty Acid Concentration in Virgin Olive Oils Using Metal Oxide Semiconductor Sensors and Multiple Linear Regression

This works aims to set up a rapid and nondestructive method to evaluate the advanced oxidation of virgin olive oils (VOOs). An electronic nose based on an array of six metal oxide semiconductor sensors was used, jointly with multiple linear regression (MLR), to predict the oxidized fatty acid (OFA) concentration in VOO samples characterized by different oxidative status. An MLR model constructed using five predictors was able to predict OFA concentration with an average validation error of 9%.