Gas sensor technology and AI: Forecasting lemon juice quality dynamics during the storage period

Abstract

Ensuring the quality of food is a critical area that directly impacts public health. The emission of Volatile Organic Compounds (VOCs), recognized as distinguishable aromas, is used for the prediction and evaluation of food quality. These compounds provide valuable data about the nature and quality of food and can serve as indicators for nutritional characteristics determination. Hence, in this study, the changes in the quality of lemon juice over a 120-day storage period were assessed using VOCs. Accordingly, an electronic nose (e-nose) equipped with 8 metal oxide sensors and chemometric methods were employed to investigate the quality changes of lemon juice during the storage period. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) models were used to visualize the qualitative changes in lemon juice samples over the storage period. Furthermore, for classifying lemon juice samples over the 120-day storage period, Support Vector Machine (SVM) and Artificial Neural Network (ANN) methods were employed. Ultimately, for predicting the pH and acidity values of lemon juice, Partial Least Squares Regression (PLSR), Principal Component Regression (PCR), and Multiple Linear Regression (MLR) methods were utilized. The results showed very high accuracy in classifying lemon juice samples during the storage period, and the constructed models could predict the pH and acidity parameters of lemon juice with high accuracy.