Rapid determination of capsaicin concentration in soybean oil by terahertz spectroscopy.

Capsaicin is the key composition of pepper and can be used as a marker of gutter oil for detection. The feasibility of rapid detection of capsaicin concentration in soybean oil was studied by terahertz spectroscopy. Genetic algorithm (GA) and principal component analysis (PCA) as the pretreatment method were used to obtain the best spectral features. Least square-support vector machine (LS-SVM), back propagation neural network (BPNN), and partial least squares (PLS) were combined with the pretreatment method to obtain the best determination model. The BPNN was combined with GA to obtain the best quantitative prediction results with the correlation coefficient of prediction (RP ), prediction root mean square error (RMSEP), the ratio of prediction to deviation (RPD), and range error ratio (RER) were 0.9309, 0.4030 µg/kg, 17.0421, and 2.4813, respectively. Furthermore, the detection limit of capsaicin could achieve 1.25 µg/kg in soybean oil and the accuracy of discrimination was up to 100% in the prediction set using the LS-SVM combined with GA pre-treatment. The results suggested that terahertz spectroscopy together with chemometric methods would be a promising technique for rapid determination of capsaicin concentration in soybean oil. Meanwhile, it is necessary to perform further experiments with real gutter oil samples before applying the method in practice. PRACTICAL APPLICATION: The combination of terahertz spectroscopy technology and chemometrics is a promising method for the rapid determination of capsaicin concentration in soybean oil with high efficiency.

Rapid and nondestructive determination of deoxynivalenol (DON) content in wheat using multispectral imaging (MSI) technology with chemometric methods.

Wheat is susceptible to contamination by deoxynivalenol (DON) which is regarded as a class III carcinogen. In this paper, a rapid and nondestructive method for DON content determination and contamination degree discrimination in wheat was developed by using a multispectral imaging (405-970 nm) system. Genetic algorithm (GA) and principal component analysis (PCA), as preprocessing methods, were used to obtain the best spectral characteristics. The determination model was established by combining preprocessing methods and chemometric methods including partial least squares (PLS), support vector machines (SVM) and back propagation neural network (BPNN). The best quantitative determination result was obtained based on GA-SVM with a correlation coefficient of prediction (Rp), root mean square error of prediction (RMSEP) and residual predictive deviation (RPD) of 0.9988, 365.3 µg kg-1 and 8.6, respectively. Furthermore, the accuracy of contamination degree classification was up to 94.29% in the prediction set by using the PCA-PLS model. The results showed that the combination of multispectral imaging technology and chemometrics was an effective and nondestructive method for the determination of DON in wheat.