Strontium isotopes in food traceability: Mechanisms, applications, and integration with machine learning

Area of Science:

• Food science and analytical chemistry.
• Geochemistry and isotopic analysis.
• Computational methods in food authentication.

Background:

• Food provenance is vital for global supply chain authenticity, safety, and quality.
• Strontium (Sr) isotopes offer a stable and geologically linked marker for food traceability.
• Existing methods face challenges with complex datasets and overlapping isotopic ranges.

Purpose of the Study:

• To review the application of Sr isotope ratios in diverse food categories.
• To highlight the stability and discriminatory power of Sr isotopes in raw and processed foods.
• To emphasize the integration of Sr isotopes with machine learning (ML) for improved food authentication.

Main Methods:

• Systematic review of scientific literature on Sr isotope applications in food.
• Analysis of Sr isotope stability and discriminatory power across various food matrices.
• Integration of Sr isotope data with ML algorithms (PCA, LDA, SVM, HCA).

Main Results:

• Sr isotope ratios demonstrate stability and discriminatory power in primary crops, livestock, and processed foods.
• ML algorithms significantly improve classification accuracy when combined with Sr isotope data.
• Synergistic use of Sr isotopes and ML provides a robust framework for food traceability.

Conclusions:

• Sr isotopes are a powerful tool for food provenance and authentication.
• Integrating Sr isotopes with ML algorithms enhances the reliability of food traceability systems.
• This approach advances scientific understanding and practical applications in global food authentication.