Nondestructive VIS/NIR spectroscopy estimation of intravitelline vitamin E and cholesterol concentration in hen shell eggs

Designer hen shell eggs, with enhanced or reduced specific components, are used as part of a healthy diet, and such designer eggs must meet certain standards. A fast, nondestructive technology is required in order to ensure that these standards are met. In this study, we investigated the nondestructive estimation of intravitelline vitamin E and cholesterol concentrations in hen shell eggs using visible and near-infrared (VIS/NIR) spectroscopy. The values estimated using partial least-squares regression (PLSR) models were highly correlated with vitamin E concentration measurements by high-performance liquid chromatography and cholesterol concentration measurements by enzymatic analysis. Wavelength selection with the Martens uncertainty test successfully removed unimportant variables and enhanced the models' accuracy and reliability. However, spectral measurements of separated egg yolks with artificially controlled vitamin E and cholesterol concentrations did not afford accurate predictive models. This result for separated yolks, together with band assignments of PLSR models for hen shell eggs, suggested that VIS/NIR spectroscopy can indirectly estimate vitamin E and cholesterol concentrations in hen shell eggs by using information about other components.