Authentication of Galician (N.W. Spain) quality brand potatoes using metal analysis. Classical pattern recognition...

Area of Science:

• Analytical Chemistry
• Chemometrics
• Food Science

Background:

• Authenticating food products with geographical indications is crucial for consumer trust and market integrity.
• Distinguishing high-quality, certified products from conventional ones requires reliable analytical methods.
• Elemental profiling offers a potential fingerprint for agricultural products.

Purpose of the Study:

• To develop a robust classification system for authenticating Galician potatoes with a Certified Brand of Origin and Quality (CBOQ).
• To differentiate CBOQ potatoes from non-CBOQ potatoes using elemental composition.
• To introduce and validate a novel vector quantization-based classification procedure (VQBCP).

Main Methods:

• Elemental analysis of 307 potato samples (CBOQ and Non-CBOQ) using atomic spectroscopy for 10 elements (K, Na, Rb, Li, Zn, Fe, Mn, Cu, Mg, Ca).
• Application of multivariate chemometric techniques including cluster analysis and principal component analysis for data exploration.
• Evaluation of supervised pattern recognition methods (LDA, KNN, SIMCA, MLF-ANN) and a newly proposed VQBCP.

Main Results:

• Elemental profiles provided sufficient information for classifying potatoes based on origin and brand.
• The proposed VQBCP achieved recognition and prediction accuracies exceeding 97%.
• VQBCP performance was superior to KNN and comparable to LDA, SIMCA, and MLF-ANN, especially in complex data structures.

Conclusions:

• Elemental analysis combined with chemometrics is effective for authenticating CBOQ Galician potatoes.
• The novel VQBCP offers a high-performance classification method for distinguishing potato categories.
• Metal profiles serve as reliable markers for verifying the origin and quality brand of potatoes.