Chemometric classification and quantification of olive oil in blends with any edible vegetable oils using FTIR-ATR and Raman spectroscopy

Rapid determination and classification of crude oils by ATR-FTIR spectroscopy and chemometric methods

Classification based on °API gravity is very important to estimate the parameters related to the extraction, purification, toxicity, and pricing of crude oils. Spectroscopy methods show some advantages over ASTM and API methods for crude oil analysis. The attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy coupled with chemometric methods has been applied as a quick and non-destructive method for crude oil analysis. In this work, a new analytical method using ATR-FTIR spectroscopy associated with chemometric methods were proposed for adressing regression and classification tasks for crude oils analysis based on °API gravity values. The designed methods are rapid, economic, and nondestructive ways in production process of oil industry. The spectral data were used for estimation of °API gravity using two approaches according to PLS-R and SVM-R algorithm, separately. The ATR-FTIR spectral data were also analyzed by classification method using the partial least squares-discriminant analysis (PLS-DA) for crude oil classification. The samples were classified into three classes based on their °API gravity values. The SVM-R model showed better results than PLS-R for °API gravity values using the F-test at 95% of confidence. The result of classification, showed about 100% accuracy and a zero classification error for calibration and prediction samples in PLS-DA algorithm.

Adaptive compressed sensing of Raman spectroscopic profiling data for discriminative tasks

Raman spectroscopy is widely used in discriminative tasks. It provides a wide-range physio-chemical fingerprint in a rapid and non-invasive way. The Raman spectrometry uses a sensor array to convert photon signals into digital spectroscopic data. This analog-to-digital process can benefit from the compressed sensing (CS) technique. The major benefits include less memory usage, shorter acquisition time, and more cost-efficient sensor. Traditional compressed sensing and reconstruction is a series of mathematical operations performed on the signal. Meanwhile, for discriminative tasks, both the signal and the categorical information are involved. For such scenarios, this paper proposes a method that uses both domain signal and categorical information to optimize CS hyper-parameters, including 1) the sampling ratio or the sensing matrix, 2) the basis matrix for the sparse transform, and 3) the regularization rate or shrinkage factor for L1-norm minimization. A case study of formula milk brand identification proves the proposed method can generate effective compressed sensing while preserving enough discriminative power in the reconstructed signal. Under the optimized hyper-parameters, a 100% classification accuracy is retained by only sampling 20% of the original signal.

Rapid Classification and Quantification of Camellia (Camellia oleifera Abel.) Oil Blended with Rapeseed Oil Using FTIR-ATR Spectroscopy

Currently, the authentication of camellia oil (CAO) has become very important due to the possible adulteration of CAO with cheaper vegetable oils such as rapeseed oil (RSO). Therefore, we report a Fourier transform infrared (FTIR) spectroscopic method for detecting the authenticity of CAO and quantifying the blended levels of RSO. In this study, two characteristic spectral bands (1119 cm^-1 and 1096 cm^-1) were selected and used for monitoring the purity of CAO. In combination with principal component analysis (PCA), linear discriminant analysis (LDA), and partial least squares regression (PLSR) analysis, qualitative and quantitative methods for the detection of camellia oil adulteration were proposed. The results showed that the calculated I₁₁₁₉/I₁₀₉₆ intensity ratio facilitated an initial check for pure CAO and six other edible oils. PCA was used on the optimized spectral region of 1800-650 cm^-1. We observed the classification of CAO and RSO as well as discrimination of CAO with RSO adulterants. LDA was utilized to classify CAO from RSO. We could differentiate and classify RSO adulterants up to 1% v/v. In the quantitative PLSR models, the plots of actual values versus predicted values exhibited high linearity. Root mean square error of calibration (RMSEC) and root mean square error of cross validation (RMSECV) values of the PLSR models were 1.4518%-3.3164% v/v and 1.7196%-3.8136% v/v, respectively. This method was successfully applied in the classification and quantification of CAO adulteration with RSO.

Non-Targeted Authentication Approach for Extra Virgin Olive Oil

The aim of this study is to develop a non-targeted approach for the authentication of extra virgin olive oil (EVOO) using vibrational spectroscopy signatures combined with pattern recognition analysis. Olive oil samples (n = 151) were grouped as EVOO, virgin olive oil (VOO)/olive oil (OO), and EVOO adulterated with vegetable oils. Spectral data was collected using a compact benchtop Raman (1064 nm) and a portable ATR-IR (5-reflections) units. Oils were characterized by their fatty acid profile, free fatty acids (FFA), peroxide value (PV), pyropheophytins (PPP), and total polar compounds (TPC) through the official methods. The soft independent model of class analogy analysis using ATR-IR spectra showed excellent sensitivity (100%) and specificity (89%) for detection of EVOO. Both techniques identified EVOO adulteration with vegetable oils, but Raman showed limited resolution detecting VOO/OO tampering. Partial least squares regression models showed excellent correlation (Rval ≥ 0.92) with reference tests and standard errors of prediction that would allow for quality control applications.

Monitoring Fatty Acid and Sterol Profile of Nizip Yaglik Olive Oil Adulterated by Cotton and Sunflower Oil

In this study virgin olive oil obtained from cv. Nizip Yaglik (NY) was adulterated with different proportions (5 and 10%, v/v) of cotton (CO) and sunflower (SO) oils. Fatty acid and sterol profiles of olive oil were analyzed by using gas chromatography (GC). Also, difference of Equivalent Carbon Number 42 values (ΔECN42) of oil samples were determined by using GC and HPLC. Due to results of fatty acids analysis, the percentage of oleic acid was decreased when CO and SO were added. Palmitic acid was increased over the addition of CO, and decreased with the addition of SO. The ΔECN42 values were increased in adulterated oils. These values showed further increase in adulterated oils with SO. Beta-sitosterols decreased to 91.06 and 88.54% when mixed with 5 and 10% SO, respectively. On the other hand, decline was negligible when mixed with CO. According to principal component analyses (PCA), pure NY and adulterated oils were clearly separated in different parts of screen plot according to fatty acids, triacylglycerol (TAGs) and sterol profile. The outcomes of this first investigation provide valuable information for about the differences of fatty acids, ΔECN42 values and sterol compounds between Turkish olive oil from Nizip Yaglik cv. and its adulteration with cotton and sunflower oil. It was observed that fatty acids are not very effective in detecting adulteration of NY oil, but ΔECN values, sterols and Rmar values can be used to detect adulteration of NY olive oil.

Alternative data mining/machine learning methods for the analytical evaluation of food quality and authenticity - A review

In recent years, the variety and volume of data acquired by modern analytical instruments in order to conduct a better authentication of food has dramatically increased. Several pattern recognition tools have been developed to deal with the large volume and complexity of available trial data. The most widely used methods are principal component analysis (PCA), partial least squares-discriminant analysis (PLS-DA), soft independent modelling by class analogy (SIMCA), k-nearest neighbours (kNN), parallel factor analysis (PARAFAC), and multivariate curve resolution-alternating least squares (MCR-ALS). Nevertheless, there are alternative data treatment methods, such as support vector machine (SVM), classification and regression tree (CART) and random forest (RF), that show a great potential and more advantages compared to conventional ones. In this paper, we explain the background of these methods and review and discuss the reported studies in which these three methods have been applied in the area of food quality and authenticity. In addition, we clarify the technical terminology used in this particular area of research.

Improved continuous locality preserving projection for quantification of extra virgin olive oil adulteration by using laser-induced fluorescence

An optimized dimensionality reduction technique is proposed as the improved continuous locality preserving projection (ICLPP), which was developed by modifying and optimizing the weighting functions and weighting factors of the continuous locality preserving projection (CLPP) algorithm. With only one adjustable parameter, this optimized technique not only enhances CLPP's capability of maintaining the continuity of the massive data, but also results in better simplicity and adaptability of the algorithm. In this paper, the performance of ICLPP is validated through quantification analysis of the adulteration of extra virgin olive oil (EVOO) with low-cost oils based on laser-induced fluorescence spectroscopy. Through cross validation and comparative studies, ICLPP, combined with the regression algorithm, is employed to predict and screen adulteration in EVOO, and is found to generally outperform other state-of-the-art dimensionality reduction algorithms, especially for prediction of adulterants at low level (<10%). It is evidenced that the ICLPP-based framework is superior in detecting adulteration by using spectral data.