Determination of glucose concentration from NIR spectra using independent component regression

In this paper, a new model based on independent component regression (ICR) is proposed to predict the glucose concentration from near infrared (NIR) spectra. The efficiency of the proposed model is validated using mixtures composed of glucose, urea and triacetin. The whole experiments were carried out in a non-controlled environment or sample conditions to show that the proposed model can suppress effectively most of the experimental variations. The proposed model decreases the standard error of prediction (SEP) from 35.59 mg/dL for Partial Least Square regression (PLS) and from 29.1 mg/dL for ICR to only 24.1 mg/dL.