On the estimation of sugars concentrations using Raman spectroscopy and artificial neural networks.

In this paper, we present an analysis of the performance of Raman spectroscopy, combined with feed-forward neural networks (FFNN), for the estimation of concentration percentages of glucose, sucrose, and fructose in water solutions. Indeed, we analysed our method for the estimation of sucrose in three solid industrialized food products: donuts, cereal, and cookies. Concentrations were estimated in two ways: using a non-linear fitting system, and using a classifier. Our experiments showed that both the classifier and the fitting systems performed better than a Support Vector Machine (SVM), a Linear Discriminant Analysis (LDA), a Linear Regression (LR), and interval Partial Least Squares (iPLS). The best-case obtained by an FFNN for water solutions was 93.33% of classification and 3.51% of Root Mean Square Error in Prediction (RMSEP), compared with 82.22% obtained by a LDA. Our proposed method got an RMSEP of 1% for the best-case obtained with the food products.

Characterization of glycated hemoglobin based on Raman spectroscopy and artificial neural networks.

The World Health Organization has declared the glycated hemoglobin (HbA1c) as a gold standard biomarker for diabetes diagnosis; this has led to relevant research on the spectral behavior and characterization of HbA1c. This paper presents an analysis of Raman peaks of commercial lyophilized HbA1c, diluted in distilled water, using concentrations of 4.76% and 9.09%, as well as pure powder (100% concentration). Vibrational Raman peak positions of HbA1c powder were found at 1578, 1571, 1536, 1436, 1311, 1308, 1230, 1222, 1114, 1106, 969, 799 and 665 cm-1; these values are consistent with results reported in other works. Besides, a nonlinear regression model based on a Feed-Forward Neural Network (FFNN) was built to quantify percentages of HbA1c for unknown concentrations. Using the Raman spectra as independent variables, the regression provided a Root Mean Square Error in Cross-Validation (RMSECV) of 0.08% ±â€¯0.04. We also include a detailed molecular assignment of the average spectra of lyophilized powder of HbA1c.