Prediction of total phenolics and flavonoids contents in Chinese wild rice (Zizania latifolia) using FT-NIR spectroscopy

Abstract

Total phenolics and flavonoids contents in Chinese wild rice were predicted using Near Infrared (NIR) spectroscopy as a rapid method. A Partial Least Square (PLS) algorithm was applied to perform the calibration. The models were calibrated by cross-validation and the chosen number of PLS factor was achieved according to the lowest Root Mean Square Error Cross-Validation (RMSECV) in calibration set. The correlation coefficient (R) and Root Mean Square of Error Prediction (RMSEP) in the test set were used as the evaluation parameters for the optimal model as follows: R = 0.985; RMSEP = 2.41 and the Residual Predictive Deviation (RPD) = 6.06 for total phenolics contents prediction by Multiplication Scatter Correction (MSC) model. For flavonoids contents prediction, R = 0.978, RMSEP = 1.23 and RPD = 4.81 by non preprocessing model. It could be concluded that NIR spectroscopy has significant potential in the nondestructive determination of total phenolics and flavonoids.