Mazarakioti Eleni C, Zotos Anastasios, Verykios Vassilios S, Kokkotos Efthymios, Thomatou Anna-Akrivi, Kontogeorgos Achilleas, Patakas Angelos, Ladavos Athanasios
Department of Food Science and Technology, University of Patras, 30131 Agrinio, Greece.
Department of Sustainable Agriculture, University of Patras, 30131 Agrinio, Greece.
Foods. 2024 Sep 23;13(18):3015. doi: 10.3390/foods13183015.
Greek giant beans, also known as "Gigantes Elefantes" (elephant beans, L.,) are a traditional and highly cherished culinary delight in Greek cuisine, contributing significantly to the economic prosperity of local producers. However, the issue of food fraud associated with these products poses substantial risks to both consumer safety and economic stability. In the present study, multi-elemental analysis combined with decision tree learning algorithms were investigated for their potential to determine the multi-elemental profile and discriminate the origin of beans collected from the two geographical areas. Ensuring the authenticity of agricultural products is increasingly crucial in the global food industry, particularly in the fight against food fraud, which poses significant risks to consumer safety and economic stability. To ascertain this, an extensive multi-elemental analysis (Ag, Al, As, B, Ba, Be, Ca, Cd, Co, Cr, Cs, Cu, Fe, Ga, Ge, K, Li, Mg, Mn, Mo, Na, Nb, Ni, P, Pb, Rb, Re, Se, Sr, Ta, Ti, Tl, U, V, W, Zn, and Zr) was performed using Inductively Coupled Plasma Mass Spectrometry (ICP-MS). Bean samples originating from Kastoria and Prespes (products with Protected Geographical Indication (PGI) status) were studied, focusing on the determination of elemental profiles or fingerprints, which are directly related to the geographical origin of the growing area. In this study, we employed a decision tree algorithm to classify Greek "Gigantes Elefantes" beans based on their multi-elemental composition, achieving high performance metrics, including an accuracy of 92.86%, sensitivity of 87.50%, and specificity of 96.88%. These results demonstrate the model's effectiveness in accurately distinguishing beans from different geographical regions based on their elemental profiles. The trained model accomplished the discrimination of Greek "Gigantes Elefantes" beans from Kastoria and Prespes, with remarkable accuracy, based on their multi-elemental composition.
希腊巨型豆，也被称为“Gigantes Elefantes”（大象豆，L.），是希腊美食中一种传统且备受珍视的烹饪佳肴，对当地生产者的经济繁荣做出了重大贡献。然而，与这些产品相关的食品欺诈问题对消费者安全和经济稳定都构成了重大风险。在本研究中，研究了多元素分析与决策树学习算法相结合用于确定多元素特征并区分从两个地理区域收集的豆子产地的潜力。在全球食品行业中，确保农产品的真实性变得越来越重要，特别是在打击食品欺诈方面，食品欺诈对消费者安全和经济稳定构成了重大风险。为了确定这一点，使用电感耦合等离子体质谱法（ICP-MS）进行了广泛的多元素分析（银、铝、砷、硼、钡、铍、钙、镉、钴、铬、铯、铜、铁、镓、锗、钾、锂、镁、锰、钼、钠、铌、镍、磷、铅、铷、铼、硒、锶、钽、钛、铊、铀、钒、钨、锌和锆）。对源自卡斯托里亚和普雷斯佩斯（具有受保护地理标志（PGI）地位的产品）的豆类样品进行了研究，重点是确定与种植地区的地理来源直接相关的元素特征或指纹。在本研究中，我们采用决策树算法根据希腊“Gigantes Elefantes”豆的多元素组成进行分类，获得了高性能指标，包括92.86%的准确率、87.50%的灵敏度和96.88%的特异性。这些结果证明了该模型基于元素特征准确区分不同地理区域豆子的有效性。训练后的模型基于其多元素组成，以极高的准确率完成了对源自卡斯托里亚和普雷斯佩斯的希腊“Gigantes Elefantes”豆的区分。
