Fecha de publicación: -- Fuente: PubMed "olive oil" Food Chem. 2026 Sep 4;528:151045. doi: 10.1016/j.foodchem.2026.151045. Online ahead of print.ABSTRACTExtra virgin olive oil (EVOO) is widely recognized as a premium grade oil, commanding a high market value and a strong reputation for health benefits. The adulteration of EVOO with low grade oils increases, especially when the adulterants are chemically similar. This study developed a detection strategy based on terahertz time domain spectroscopy (THz-TDS) for detecting adulteration in EVOO. Four datasets were constructed by mixing two brands of extra virgin olive oil with two like adulterants (expired olive oil and refined olive oil) with concentrations ranging from 5% to 50% and obtaining the terahertz absorption spectra of each sample; non-targeted discrimination was performed using Principal Component Analysis (PCA) coupled with Data-Driven Soft Independent Analogical Modeling (DD-SIMCA) in order to differentiate between the normal and adulterated samples. For quantitative analysis, a partial least squares regression (PLSR) model based on two variable selection methods (Competitive Adaptive Reweighted Sampling, CARS, and Variable Combination Population Analysis, VCPA) was developed to accurately predict the specific adulteration concentration in the samples. The results demonstrated that DD-SIMCA achieved 100% correct classification of authentic samples. Furthermore, the PLSR models exhibited excellent predictive performance across all datasets, with coefficient of determination (R2) values exceeding 0.95, residual predictive deviation (RPD) values greater than 5, and root mean square error of prediction (RMSEP) below 3.5%. Meanwhile, the predictive performance of the optimal model on external blind samples also reaches 0.94. The results show that THz-TDS is a very promising tool for non-destructive and rapid assessment of the authenticity of EVOO.PMID:42705073 | DOI:10.1016/j.foodchem.2026.151045