The sale of frozen–thawed fish and fish products, labeled as fresh, is currently one of the most common and insidious commercial food frauds. For this reason, the demand of reliable tools to identify the storage conditions is increasing. The present study was performed on two species, commonly sold in large-scale distribution: Cuttlefish (Sepia officinalis) and musky octopus (Eledone spp.). Fifty fresh cephalopod specimens were analyzed at refrigeration temperature (2 ± 2 °C), then frozen at −20 °C for 10 days and finally thawed and analyzed again. The performance of three near-infrared (NIR) instruments in identifying storage conditions were compared: The benchtop NIR Multi Purpose Analyzer (MPA) by Bruker, the portable MicroNIR by VIAVI and the handheld NIR SCiO by Consumer Physics. All collected spectra were processed and analyzed with chemometric methods. The SCiO data were also analyzed using the analytical tools available in the online application provided by the manufacturer to evaluate its performance. NIR spectroscopy, coupled with chemometrics, allowed discriminating between fresh and thawed samples with high accuracy: Cuttlefish between 82.3–94.1%, musky octopus between 91.2–97.1%, global model between 86.8–95.6%. Results show how food frauds could be detected directly in the marketplace, through small, ultra-fast and simplified handheld devices, whereas official control laboratories could use benchtop analytical instruments, coupled with chemometric approaches, to develop accurate and validated methods, suitable for regulatory purposes.

Fish species substitution is one of the most common forms of fraud all over the world, as fish identification can be very challenging for both consumers and experienced inspectors in the case of fish sold as fillets. The difficulties in distinguishing among different species may generate a “grey area” in which mislabelling can occur. Thus, the development of fast and reliable tools able to detect such frauds in the field is of crucial importance. In this study, we focused on the distinction between two flatfish species largely available on the market, namely the Guinean sole (Synaptura cadenati) and European plaice (Pleuronectes platessa), which are very similar looking. Fifty fillets of each species were analysed using three near-infrared (NIR) instruments: the handheld SCiO (Consumer Physics), the portable MicroNIR (VIAVI), and the benchtop MPA (Bruker). PLS-DA classification models were built using the spectral datasets, and all three instruments provided very good results, showing high accuracy: 94.1% for the SCiO and MicroNIR portable instruments, and 90.1% for the MPA benchtop spectrometer. The good classification results of the approach combining NIR spectroscopy, and simple chemometric classification methods suggest great applicability directly in the context of real-world marketplaces, as well as in official control plans.

Hazelnuts (Corylus avellana L.) are among the most consumed dry fruits all over the world. Their commercial quality is defined, above all, by origin and dimension, as well as by lipid content. Evaluation of this parameter is currently performed with chemical methods, which are expensive, time consuming, and complex. In the present work, the near-infrared (NIR) spectroscopy, using both a benchtop research spectrometer and a retail handheld instrument, was evaluated in comparison with the traditional chemical approach. The lipid content of hazelnuts from different growing regions of origin (Italy, Chile, Turkey, Georgia, and Azerbaijan) was determined with two NIR instruments: a benchtop FT-NIR spectrometer (Multi Purpose Analyser—MPA, by Bruker), equipped with an integrating sphere and an optic fibre probe, and the pocket-sized, battery-powered SCiO molecular sensor (by Consumer Physics). The Randall/Soxtec method was used as the reference measurement of total lipid content. The collected NIR spectra were inspected through multivariate data analysis. First, a Principal Component Analysis (PCA) model was built to explore the information contained in the spectral datasets. Then, a Partial Least Square (PLS) regression model was developed to predict the percentage of lipid content. PCA showed samples distributions that could be linked to their total crude fat content determined with the Randall/Soxtec method, confirming that a trend related to the lipid content could be detected in the spectral data, based on their chemical profiles. PLS models performed better with the MPA instrument than SCiO, with the highest R² of prediction (R²_PRED = 0.897) achieved by MPA probe, while this parameter for SCiO was much lower (R²_PRED = 0.550). Further analyses are necessary to evaluate if more acquisitions may lead to better performances when using the SCiO portable spectrometer.