Fish: Freshness & Freeze Thaw Detection

Fish: Freshness & Freeze Thaw Detection

Sardina

1. The Problem

Fish is among the most perishable commodities on the market, and freezing is used throughout the supply chain to extend its marketability. However, previously frozen fish is frequently sold as fresh to capture the price difference between the two product categories. This practice constitutes economic fraud and undermines consumer confidence.1,2,3

European legislation (Regulation 1169/2011) requires that previously frozen fish sold as fresh be clearly labelled, yet enforcement is hampered by the lack of rapid, field deployable detection tools.

The current techniques (K value, TVB N, microbiological counts, low field NMR, MRI and calorimetry) are laboratory based, slow, often destructive, and they reflect general spoilage and postmortem storage rather than the freeze and thaw history itself, so repeated freeze thaw cycles go undetected.

2. The Production Chain

Every step of the fish value chain offers an opportunity for rapid, non destructive NIR screening:

  • 1. Capture or aquaculture and landing at the port: the product is iced or frozen on board and inspected at landing.
  • 2. Freezing, cold storage and processing: freezing on board or at the plant, where freeze thaw history matters most and portable NIR screening can be applied directly on the fish at the port or plant.
  • 3. Importer, wholesaler and distribution: imported consignments can be verified in minutes to confirm that what is received matches what is declared, before redistribution.
  • 4. Retail and Horeca: quality control before sale protects the reputation of fishmongers, restaurants and supermarkets.
  • 5. Consumer: benefits indirectly through confidence that the product labelled as fresh truly is fresh.

3. Where NIR Can Contribute

In our study, conducted on Atlantic sardines (Sardina pilchardus), a portable NIR spectrometer operating from 908 to 1676 nm was used to measure 188 sardines and 1,831 spectra across seven storage classes: fresh, frozen cycle 1, frozen cycle 2, refrigerated cycle 1, refrigerated cycle 2, thawed cycle 1 and thawed cycle 2. In our particular study a Viavi MicroNIR OnSite-W instrument was used, though any portable NIR spectrometer covering a similar range can be used.

Hierarchical cascade98.6%all seven classes
Individual models100%binary and ternary
Freeze Thaw IndexR² 0.905global, holdout

Beyond classification, a continuous Freeze Thaw Index was built from two mechanistically interpretable sub indices: a Water and Hydration index (1330 to 1530 nm) and a Lipid Oxidation index (920 to 960 and 1620 to 1676 nm), both rising monotonically from fresh to twice thawed sardines on an independent validation set.

The index was packaged into a standalone graphical interface that displays a colour coded percentage (green for fresh, yellow for partial, red for heavily freeze thawed) together with a plain language status label, so inspectors and seafood businesses can obtain a result in seconds without any chemometric expertise.

4. Other Applications: Contaminants and Safety

The same point measurement principle extends to safety screening. Published studies have explored:

  • Heavy metals: near infrared reflectance spectroscopy has been used to detect heavy metal contamination in mussels, classifying contaminated samples rapidly and without reagents.4
  • Histamine (toxin): NIR combined with machine learning has been applied to evaluate histamine contamination in frozen thawed tuna, with good quantification and high classification accuracy.5
  • Urea contamination: NIR absorbance data and machine learning classified fish samples as safe or unsafe against a 1000 ppm urea threshold, without sample preparation.6

Near infrared spectroscopy is therefore positioned as a rapid, non destructive and cost efficient tool that extends from freshness and freeze thaw authentication to the emerging field of safety screening in fish and fishery products.7

5. References

  1. Atanassova S, Yorgov D. Near-Infrared Spectroscopy for Rapid Differentiation of Fresh and Frozen-Thawed Common Carp (Cyprinus carpio). Sensors. 2024. Available from: Sensors 2024
  2. Reis MM, et al. Non-invasive differentiation between fresh and frozen/thawed tuna fillets using near infrared spectroscopy (Vis-NIRS). LWT. 2017. Available from: LWT 2017
  3. Uddin M, et al. Non-destructive Visible/NIR Spectroscopy for Differentiation of Fresh and Frozen-thawed Fish. Journal of Food Science. 2005. Available from: J Food Sci 2005
  4. Regulation (EU) No 1169/2011 on the provision of food information to consumers. EUR-Lex. Available from: eur-lex.europa.eu
  5. Liu Y, et al. Rapid detection of mussels contaminated by heavy metals using near-infrared reflectance spectroscopy and a constrained difference extreme learning machine. Spectrochimica Acta Part A. 2022. Available from: Spectrochim Acta A 2022
  6. Curro S, et al. Application of near-infrared spectroscopy as at-line method for the evaluation of histamine in tuna fish (Thunnus albacares). Food Control. 2025. Available from: Food Control 2025
  7. Ninh DK, et al. Classification of Urea Content in Fish Using Absorbance Near-Infrared Spectroscopy and Machine Learning. Applied Sciences. 2024. Available from: Applied Sciences 2024
  8. Zhou J, et al. Applications of Near-Infrared Spectroscopy for Nondestructive Quality Analysis of Fish and Fishery Products. Foods. 2024. Available from: Foods 2024