Visualization of volatomic profiles for early detection of fungal infection on storage Jasmine brown rice using electronic nose coupled with chemometrics

Abstract

Early detection of fungal contamination can prevent fungal infected rice to enter food chains. This study aimed to use electronic nose coupled with chemometrics as a rapid and nondestructive method for detection of fungal contamination on brown rice grain. Jasmine brown rice was artificially infected with Aspergillus and stored at 30 °C and 85% RH. Volatile markers of fungal infected brown rice were identified using solid phase microextraction/gas chromatography–mass spectrometry (SPME/GC–MS). The volatomic profiles of fungal infected rice were analyzed using electronic nose and explored using principal component analysis (PCA). Linear discriminant analysis (LDA) and support vector machine (SVM) were then employed to classify samples with different levels of fungal contamination as a factor of storage time. Partial least squares (PLS) regression model was developed for prediction of the fungal growth on brown rice. The electronic nose coupled with PLS regression could accurately predict the fungal growth and gave coefficient of determination, R2 = 0.969, and root mean square error, RMSE = 0.31 Log CFU/g. The results suggested that the electronic nose can be used as a rapid and nondestructive tool for early detection of fungal infection on rice grain prior to visible growth.