plsr (3)

8198791899?profile=RESIZE_400x

Pistachio is one of the most expensive nuts, and is prone to adulteration because of its high commodity value. The most common adulterants are green pea and peanuts with added colours. Turkish researchers have developed a non-targeted method using portable FT-IR (Fourier Transform infared) and UV–Visible spectrometers.  Samples of pistachio granules were adulterated with green pea and peanut at concentrations from 5-40% w/w, and their spectra taken using  a portable FT-IR spectrometer and a conventional UV–Vis spectrometer, which were analysed by Soft Independent Modeling of Class Analogy (SIMCA) to generate classification algorithms to authenticate pistachio. Partial Least Square Regression (PLSR) was used to predict the concentrations of adulterants, and both instruments gave excellent predictions of adulterant levels.

Read the abstract here

Read more…

4248264363?profile=RESIZE_710xBecause of its higher price, goats milk is vulnerable to adulteration by cheaper cows milk. A rapid method based on β-carotene as a marker for cows milk, but absent in goats milk, has been developed using Raman and Infrared spectroscopy with chemometrics. The application of PLSR (partial least squares regression) to the two spectrocopic methods was the most successful giving an good correlation coefficient of validation (R2 value > 96) and an accurate determination of β-carotene content and percentage in spiked milk.

Read the abstract here

Read more…

Urea is added as an adulterant to give milk whiteness and increase its consistency by improving the non-fat solids content, but excessive amounts of urea in milk causes overload and kidney damage. A sensitive method for detecting and quantifying urea adulteration of milk has been developed using FT-NIRS (Fourier Transformed Near Infra Red Spectroscopy) coupled with multivariate analysis. The model was developed using 162 fresh milk samples, consisting of 20 non-adulterated samples (without urea), and 142 samples with the urea adulterant at 8 different concentrations (0.10%, 0.30%, 0.50%, 0.70%, 0.90%, 1.10%, 1.30%, and 1.70%), each prepared in triplicate. The NIR data coupled with the PLS‐DA (Partial Least Squares -Discriminant Analysis) model can be used to discriminate between the unadulterated fresh milk samples and those adulterated with urea.  Furthermore, the NIR data coupled with PLSR (Partial Least Squares Regression) models may be used to quantify the level of the urea in milk samples. 

2018399?profile=RESIZE_710x Read the full paper

Read more…