Predicting single kernel and bulk milled rice alkali spreading value and gelatinization temperature class using NIR spectroscopy

Background and Objectives The alkali spreading value (ASV) of rice is a widely measured quality parameter and accepted indicator of gelatinization temperature (GT) class. However, the alkali test, developed in 1958, is labor intensive and subjective. Better methods to measure ASV and GT, for single kernel and bulk rice would provide an important tool to determine the effects of individual kernels on end-use quality and rice with desired cooking qualities. An instrument developed by the USDA-ARS and a commercially available near infrared (NIR) instrument were evaluated for determining ASV and classification of intermediate and low GT levels for single kernel and bulk milled rice, respectively. Findings Quantitative prediction of ASV scores (2–7) demonstrated the potential of NIR spectroscopy for screening with a standard error of prediction for validation samples ranging from 0.91 to 1.39 for the single kernel NIR instrument, and from 0.97 to 1.19 for the commercial instrument. GT categorization into intermediate and low values, using ASV scores, showed 82.4% and 85.0% correct classification using 1 and 30-single kernel average calibration models, respectively. GT was correctly classified (93.6%–84.4%) using a commercial NIR instrument. Conclusion NIR spectroscopy has potential for rough screening of ASV and for two-category GT rice classification. Significance and Novelty Considering that NIR spectroscopy has been proven to be applicable for other quality parameters, such as rice starch content and quality, protein content, and milling degree, the addition of calibrations for ASV as a predictor of GT class will be highly beneficial while not requiring additional resources. The rapid and nondestructive classification of individual kernels may also enable physical segregation of individual kernels for use by rice researchers and/or industry for additional studies on kernels with specific quality parameters and for determining the extent of variant kernels in a milled rice lot that could affect end use quality. Future studies on the use of NIR spectroscopy for brown rice should be evaluated so that quality could be assessed, while maintaining seed viability, and then used in field studies.