**4. Conclusions**

This study demonstrates the possibility of using a handheld NIR spectrometer as an alternative to GC-FID to determine fish oil fat composition on-site in a fast and nondestructive way. NIR spectroscopy, coupled with chemometrics, can predict concentrations of SFAs, MUFAs, PUFAs and ω-3 FAs with good results, with the SFAs and ω-3 models performing best in external validation (R<sup>2</sup> of 0.98 and 0.99, RMSEP = 0.94% and 0.98%, and BIAS = 0.78% and −0.67%, respectively, in the test set).

Although the technique produced a high error of prediction and bias in the ω-6 FAs model (RMSEP = 2.09% and Bias = −1.76%), this was corrected with the application of BSC, obtaining an R<sup>2</sup> of 0.95, an RMSEP of 1.09% and a bias of −0.05%, which could be improved in the future with the addition of new oil samples to the model.

The results presented in this study demonstrate that NIR spectroscopy is a mature technology capable of rapidly and efficiently determining the quality of oils extracted from fish by-products, which makes it suitable for industrial applications. This will allow food industries to rapidly and efficiently determine the quality and commercial value of oil coming from fish by-products.

**Author Contributions:** Conceptualization, I.O., E.S. and Á.M.-H.; methodology, S.N.-O., I.O., E.S. and Á.M.-H.; validation, S.N.-O., I.O., E.S. and Á.M.-H.; formal analysis, S.N.-O. and Á.M.-H.; investigation, S.N.-O., I.O., E.S. and Á.M.-H.; data curation, S.N.-O., E.S. and Á.M.-H.; writing—original draft preparation, S.N.-O., I.O., G.A., G.F. and Á.M.-H.; writing—review and editing, S.N.-O., I.O., G.A., G.F. and Á.M.-H.; visualization, S.N.-O., I.O. and Á.M.-H.; project administration, I.O.; funding acquisition, I.O. and Á.M.-H. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research part of the "Gaitik -Monitoring of quality parameters for process automation: PAT technology for the improvement of production systems" project, funded by the Basque Government—Department of Economic Development, Sustainability and Environment—Vice. Dept. of Agriculture, Fishing and Food Policy, Directorate of Quality and Food Industries.

**Data Availability Statement:** The datasets generated for this study are available on request from the corresponding author.

**Acknowledgments:** The authors greatly acknowledge the Basque Government—Department of Economic Development, Sustainability and Environment.—Vice. Dept. of Agriculture, Fishing and Food Policy, Directorate of Quality and Food Industries for the funding of the project and for the scholarship of Sonia Nieto-Ortega. They also acknowledge the company Barna (Mundaka, Spain) for providing fish oil samples for this research project. This paper is contribution n◦ 1096 from AZTI, Food Research, Basque Research and Technology Alliance (BRTA).

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
