Abstract
Near-infrared reflectance spectroscopy (NIRS) was applied to develop a conventional method for estimating the glycemic index (GI) and starch digestion pattern with α-amylase of brown-rice flour. The digestion pattern was expressed by a non-linear equation including 3 parameters, "parameter a" that expresses the digestible fraction by prepared heat treatment, "parameter b" t hat expresses the amount of digestible fraction during digestion trials, and "parameter c" that expresses the starch hydrolysis rate. The GI value was expressed by the EGI obtained from the in vitro starch digestion trial. The accuracy of the validation was evaluated by the evaluation index and the ratio of standard deviation of reference data in prediction sample set to the standard error of prediction. The results of validation showed that the estimation method by NIRS was useful to estimate the value of "parameter c" and EGI and for primary screening of the varieties in the rice breeding system, suggesting that it was useful to select the rice varieties having a lower rate of starch digestion. The results also suggested that the sample segregation by the rice subspecies or production region was useful to improve the accuracy of estimation.
