Food Hygiene and Safety, Meat and Meat Products Research Institute (IProCar), Faculty of Veterinary Science, University of Extremadura, 10003 Cáceres, Spain.
Department of Food, Environmental, and Nutritional Sciences (DeFENS), Università degli Studi di Milano, Via G. Celoria 2, 20133 Milan, Italy.
Toxins (Basel). 2021 Sep 3;13(9):620. doi: 10.3390/toxins13090620.
The ripening process of dry-cured meat products is characterised by the development of fungi on the product's surface. This population plays a beneficial role, but, uncontrolled moulds represent a health risk, since some of them may produce mycotoxins, such as ochratoxin A (OTA). The aim of the present work is to assess the potential of near-infrared spectroscopy (NIRS) for the detection of OTA-producing mould species on dry-cured ham-based agar. The collected spectra were used to develop Support Vector Machines-Discriminant Analysis (SVM-DA) models by a hierarchical approach. Firstly, an SVM-DA model was tested to discriminate OTA and non-OTA producers; then, two models were tested to discriminate species among the OTA producers and the non-OTA producers. OTA and non-OTA-producing moulds were discriminated with 85% sensitivity and 86% specificity in the prediction. Furthermore, the SVM-DA model could differentiate non-OTA-producing species with a 95% sensitivity and specificity. Promising results were obtained for the prediction of the four OTA-producing species tested, with a 69% and 90% sensitivity and specificity, respectively. The preliminary approach demonstrated the high potential of NIR spectroscopy, coupled with Chemometrics, to be used as a real-time automated routine monitorization of dry-cured ham surfaces.
干腌肉制品的成熟过程的特点是产品表面真菌的生长。该菌群起着有益的作用，但不受控制的霉菌代表健康风险，因为其中一些霉菌可能会产生霉菌毒素，如赭曲霉毒素 A（OTA）。本工作旨在评估近红外光谱（NIRS）在干腌火腿琼脂上检测产 OTA 霉菌种的潜力。收集的光谱用于通过分层方法开发支持向量机-判别分析（SVM-DA）模型。首先，测试了 SVM-DA 模型以区分 OTA 和非 OTA 生产者；然后，测试了两个模型以区分 OTA 生产者和非 OTA 生产者中的物种。OTA 和非 OTA 产生的霉菌在预测中具有 85%的灵敏度和 86%的特异性。此外，SVM-DA 模型可以区分非 OTA 产生的物种，灵敏度和特异性分别为 95%。对于测试的四种产 OTA 物种的预测，获得了有希望的结果，灵敏度和特异性分别为 69%和 90%。初步方法证明了 NIR 光谱结合化学计量学作为干腌火腿表面实时自动常规监测的高潜力。
