Hyperspectral imaging and machine learning in food microbiology: Developments and challenges in detection of bacterial,fungal, and viral contaminants

Hyperspectral imaging (HSI) is a robust and nondestructive method that can detect foreign particles such as microbial, chemical, and physical contamination in food. This review summarizes the work done in the last two decades in this field with a highlight on challenges, risks, and research gaps. Considering the challenges of using HSI on complex matrices like food (e.g., the confounding and masking effects of background signals), application of machine learning and modeling approaches that have been successful in achieving better accuracy as well as increasing the detection limit have also been discussed here. Foodborne microbial contaminants such as bacteria, fungi, viruses, yeast, and protozoa are of interest and concern to food manufacturers due to the potential risk of either food poisoning or food spoilage. Detection of these contaminants using fast and efficient methods would not only prevent outbreaks and recalls but will also increase consumer acceptance and demand for shelf-stable food products. The conventional culture-based methods for microbial detection are time and labor-intensive, whereas hyperspectral imaging (HSI) is robust, nondestructive with minimum sample preparation, and has gained significant attention due to its rapid approach to detection of microbial contaminants. This review is a comprehensive summary of the detection of bacterial, viral, and fungal contaminants in food with detailed emphasis on the specific modeling and datamining approaches used to overcome the specific challenges associated with background and data complexity.     

Sensing of Mycotoxin Producing Fungi in the Processing of Grains

Due to the growing demand for rapid, sensitive and inexpensive detection of contaminations in food and feed, both in the field and after harvest, non-invasive and non-destructive spectroscopic methods are reviewed for the direct identification of fungi and mycotoxins on grain and in grain products. Different model samples containing defined contaminations of fungi and/or mycotoxins as well as real-world samples were investigated using spectroscopic methods (steady-state and time-resolved fluorescence spectroscopy, ultraviolet/visible/near infrared (NIR) spectroscopy, two-photon-induced fluorescence (TPIF)). To evaluate the capability of TPIF for on-line and in-line mycotoxin analysis in real-world samples, fundamental parameters of different mycotoxins, such as two-photon cross-sections, were determined in beer and wine. A comparison showed that the disturbance due to background emission originating from matrix constituents is significantly reduced under two-photon excitation conditions, and TPIF can be applied for a qualitative and quantitative analysis of mycotoxins in wine and beer. The combination of data obtained from different spectroscopic methods (such as optimised excitation and emission wavelengths, fluorescence decay times and fluorescence quantum efficiency on the one hand side and NIR spectroscopy on the other side) is promising for the qualitative as well as quantitative identification of fungi and mycotoxins. Moreover, NIR reflection spectra yield additional information on ingredients, moisture content and the presence (or absence) of fungi in the sample. The spectroscopic techniques are complemented by chemometric tools to extract the desired chemical information, e.g. with respect to the presence of contaminations.     

Contamination with storage fungi of human food from Cameroon

In a mycological study, a total of 95 human food samples were investigated to evaluate the incidence of fungal contamination in Cameroon by conventional identification method and partly confirmed by DNA sequencing. The isolated fungal spp. were further studied to determine their toxigenic potentials. The investigation revealed the predominance of Aspergillus and Penicillium with 96% of samples contaminated with at least one species of these fungi, whereas the incidence of co-contamination of samples was 85%. Aspergillus flavus and Aspergillus parasiticus (Flavi section) were the most predominant species contaminating mainly maize and peanuts. In addition, P. crustosum and P. polonicum were the most common contaminants belonging to the genus Penicillium. On the other hand, A. ochraceus (Circumdati section) registered a low incidence rate of 5%, including other members of the Aspergillus group. Other members of the genera Rhizopus and Alternaria spp. were also registered in the study. A majority of fungal strains of A. ochraceus, A. parasiticus, P. crustosum and P. polonicum isolated were toxigenic, producing the mycotoxins tested for, while none was detected in cultures of A. fumigatus. The high incidence rate of fungi contamination coupled with their potentials in producing mycotoxins gives a strong indication that the samples tested may likely be contaminated with various mycotoxins. There is need for further study to assess the incidence of mycotoxins contamination in similar food samples.     

PCR detection of Alternaria spp. in processed foods, based on the internal transcribed spacer genetic marker

The genus Alternaria is considered one of the most important fungal contaminants of vegetables, fruits, and cereals, producing several mycotoxins that can withstand food processing methods. Conventional methods for Alternaria identification and enumeration are laborious and time-consuming, and they might not detect toxigenic molds inactivated by food processing. In this study, a PCR method has been developed for the rapid identification of Alternaria spp. DNA in foodstuffs, based on oligonucleotide primers targeting the internal transcribed spacer (ITS) 1 and ITS2 regions of the rRNA gene. The specificity of the Alternaria-specific primer pair designed (Dir1ITSAlt-Inv1ITSAlt) was verified by PCR analysis of DNA from various Alternaria spp., and also from several fungal, bacterial, yeast, animal, and plant species. The detection limit of the method was 10(2) CFU/ml in viable culture, heated culture, or experimentally inoculated tomato pulp. The applicability of the method for detection of Alternaria spp. DNA in foodstuffs was assessed by testing several commercial samples. Alternaria DNA was detected in 100% of spoiled tomato samples, 8% of tomato products, and 36.4% of cereal-based infant food samples analyzed.     

DNA arrays, electronic noses and tongues, biosensors and receptors for rapid detection of toxigenic fungi and mycotoxins: A review

This paper presents an overview of how microsystem technology tools can be applied to the development of rapid, out-of-laboratory measurement capabilities for the determinations of toxigenic fungi and mycotoxins in foodstuffs. Most of the topics discussed are all under investigation within the European Commission-sponsored project Good-Food (FP6-IST). These are DNA arrays, electronic noses and electronic tongues for the detection of fungal contaminants in feed, and biosensors and chemical sensors based on microfabricated electrode systems, antibodies and novel synthetic receptors for the detection of specific mycotoxins. The approach to resolution of these difficult measurement problems in real matrices requires a multidisciplinary approach. The technology tools discussed can provide a route to the rapid, on-site generation of data that can aid the safe production of high-quality foodstuffs.     

Recent progress in mycotoxins detection based on surface-enhanced Raman spectroscopy

Mycotoxins are toxic compounds naturally produced by certain types of fungi. The contamination of mycotoxins can occur on numerous foodstuffs, including cereals, nuts, fruits, and spices, and pose a major threat to humans and animals by causing acute and chronic toxic effects. In this regard, reliable techniques for accurate and sensitive detection of mycotoxins in agricultural products and food samples are urgently needed. As an advanced analytical tool, surface-enhanced Raman spectroscopy (SERS), presents several major advantages, such as ultrahigh sensitivity, rapid detection, fingerprint-type information, and miniaturized equipment. Benefiting from these merits, rapid growth has been observed under the topic of SERS-based mycotoxin detection. This review provides a comprehensive overview of the recent achievements in this area. The progress of SERS-based label-free detection, aptasensor, and immunosensor, as well as SERS combined with other techniques, has been summarized, and in-depth discussion of the remaining challenges has been provided, in order to inspire future development of translating the techniques invented in scientific laboratories into easy-to-operate analytic platforms for rapid detection of mycotoxins.     

拉曼光谱在食品安全检测中的应用

食品污染物引发的食品安全问题是全球公共卫生面临的主要挑战之一。高效液相色谱、气相色谱和质谱等传统检测方法具有高准确性和灵敏度,但是这些方法需要复杂的仪器和专业技术人员并且过程耗时。因此,开发灵敏、快速、简便和无损的分析检测方法对保障食品安全具有重要意义。拉曼光谱作为一种新兴检测技术,在食品污染物检测领域引起了广泛关注。为推动拉曼光谱技术在食品安全应用中发挥更大潜力,本文从拉曼光谱技术发展的角度（从传统拉曼到表面增强空间偏移拉曼）综述了该技术在食品污染物检测方面的最新和代表性应用。最后，讨论了拉曼技术在食品安全检测中的挑战和未来前景，为食品污染物的检测和控制提供新思路。     

粮食污染物的快速检测技术研究进展

我国粮食安全的主要污染物包括真菌毒素、重金属和农药残留, 每年由此损失的粮食达几千万吨。为保障人民群众“舌尖上的安全”, 满足我国粮食收购、监管的实际需求, 建立和发展检测粮食中有毒有害物质的有效快速检测技术非常必要。本文系统地介绍了国内外粮食污染物的主要快速检测技术, 包括免疫检测技术、电化学分析方法、X-射线荧光光谱技术、酶抑制法和表面增强拉曼光谱法等。通过对这些快检方法应用及优缺点分析, 以期能为我国粮食污染物的有效监管及检测技术的发展方向提供有价值的参考。     

Potential Use of Fourier Transform Infrared Spectroscopy for Identification of Molds Capable of Producing Mycotoxins

Fourier transform infrared spectroscopy is considered as a rapid, non-destructive, reliable, sensitive, and cost-effective technique, which could be used for characterizing the chemical composition (identifying functional groups) of various microorganisms. In the present study, Fourier transform infrared spectroscopy was employed to differentiate (based on functional groups; spectral range between 3500–500 cm^?1) between different genera of fungi (mainly Aspergillus sp. and Mucor sp.), capable of producing mycotoxins. From the results, it was clearly evident that irrespective of the overall similarities between the spectra of different fungi, there is a unique spectrum for each one with precise differences in the functional groups. These results can be useful and provide a suitable “finger print” for each of the fungal strains capable of producing mycotoxins. This method of differentiating between toxigenic fungi can also be effectively used in assisting quarantine rules. The results obtained in this study revealed Fourier transform infrared spectroscopy to show great potential for identification of molds capable of producing mycotoxins.     

Current Immunoassay Methods for the Rapid Detection of Aflatoxin in Milk and Dairy Products

The presence of mycotoxins in foodstuff causes serious health problems to consumers and economically affects the food industry. Among the mycotoxins, aflatoxins are very toxic and highly carcinogenic contaminants which affect the safety of many foods, and therefore endanger human health. Aflatoxin M₁ (AFM₁) found in milk results from the biotransformation of aflatoxin B₁. Many efforts have been made to control the source of AFM₁ from farmers to dairy product companies. However, AFM₁ escapes ordinary methods of food treatment such as cooking, sterilization, and freezing, hence it appears in milk and dairy products. The presence of high levels of AFM₁ constitutes an alarming threat as milk and dairy products contain essential nutrients for human health, especially for infants and children. For this reason, there is a pressing need for developing a fast and reliable screening method for detecting trace aflatoxins in food. Several analytical methods based on high‐performance liquid chromatography (HPLC) and mass spectroscopy have been used for aflatoxin detection; however, they are expensive, time‐consuming, and require many skills. Recently, immunoassay methods, including enzyme‐linked immunosorbent assay (ELISA), immunosensors, and lateral flow immunoassay (LFIA), have been preferred for food analysis because of their improved qualities such as high sensitivity, simplicity, and capability of onsite monitoring. This paper reviews the new developments and applications of immunoassays for the rapid detection of AFM₁ in milk.     

果蔬中常见真菌毒素的检测研究进展

果蔬在田间的生长过程以及采摘后的贮存、运输和加工过程中,容易受到病原真菌的污染,病原菌通过寄主果蔬表面伤口或自然孔口侵入,导致果蔬腐烂并积累大量真菌毒素。腐烂果蔬会造成巨大的经济损失,同时经由食物链的传递威胁人和动物的健康。鉴于果蔬中真菌毒素的危害性,本文总结了近年来主要污染果蔬的毒素的种类、污染现状、危害及其检测方法,其中目前检测仍以色谱法、质谱法等仪器检测为主,虽然可以精确定量检测,但是耗材和时间成本较大;随着大规模样本的筛选检测需求的提高,免疫法和表面增强拉曼光谱法等高通量的快速检测逐渐受到关注。本文旨在为真菌毒素的检测防控提供参考。     

Research advancements in optical imaging and spectroscopic techniques for nondestructive detection of mold infection and mycotoxins in cereal grains and nuts

Cereal grains and nuts are represented as the economic backbone of many developed and developing countries. Kernels of cereal grains and nuts are prone to mold infection under high relative humidity and suitable temperature conditions in the field as well as storage conditions. Health risks caused by molds and their molecular metabolite mycotoxins are, therefore, important topics to investigate. Strict regulations have been developed by international trade regulatory bodies for the detection of mold growth and mycotoxin contamination across the food chain starting from the harvest to storage and consumption. Molds and aflatoxins are not evenly distributed over the bulk of grains, thus appropriate sampling for detection and quantification is crucial. Existing reference methods for mold and mycotoxin detection are destructive in nature as well as involve skilled labor and hazardous chemicals. Also, these methods cannot be used for inline sorting of the infected kernels. Thus, analytical methods have been extensively researched to develop the one that is more practical to be used in commercial detection and sorting processes. Among various analytical techniques, optical imaging and spectroscopic techniques are attracting growers’ attention for their potential of nondestructive and rapid inline identification and quantification of molds and mycotoxins in various food products. This review summarizes the recent application of rapid and nondestructive optical imaging and spectroscopic techniques, including digital color imaging, X-ray imaging, near-infrared spectroscopy, fluorescent, multispectral, and hyperspectral imaging. Advance chemometric techniques to identify very low-level mold growth and mycotoxin contamination are also discussed. Benefits, limitations, and challenges of deploying these techniques in practice are also presented in this paper.     

Molecular Detection of Mycotoxigenic Fungi in Foods: The Case for Using PCR-DGGE

Among the toxin-producing microbes, those that produce mycotoxins are especially problematic due to their broad distribution in the environments and in foods. Several species of Aspergillus, Penicillium, and Fusarium are sources of potent mycotoxins such as aflatoxins, ochratoxins, patulin, deoxynivalenol, and fumonisins. It is, therefore, vital that mycotoxigenic fungi contaminants in food are rapidly and accurately identified for ensuring the safety of consumers. Most of the current methods are based on PCR using gene-specific or species-specific primers. However, contaminating microbes often compose a complex community and PCR-DGGE may provide a better approach than traditional single-gene and/or single-species based methods. It provides “fingerprints” for each microbial flora and has been widely used to analyze environmental and food-associated microbial communities. This review shows the advantages and disadvantages of different molecular methods for the detection of mycotoxigenic fungi including PCR-DGGE as a potent and applicable method that could overcome the difficulties associated with other methods.     

微芯片技术在农产品及食品真菌毒素快速检测中的应用

真菌毒素污染的农产品及食品会对人类及牲畜的健康产生严重威胁。目前,真菌毒素的检测主要采用液相色谱-质谱联用法(liquid chromatography-mass spectrometry,LC-MS)、气相色谱-质谱联用法(gas chromatography-mass spectrometry,GC-MS)等传统检测方法。然而,这些检测方法的样品前处理过程比较繁琐、费时,并且在分析过程中还会消耗大量的有毒试剂。微芯片技术所需样品的消耗量少并且分析时间短,可实现样品的集成化、微型化以及高通量检测。微芯片技术在真菌毒素检测中的应用,弥补了上述传统检测方法的不足。本文主要综述了微芯片技术在农产品及食品真菌毒素快速检测中的应用研究进展。首先对农产品及食品中常见的真菌毒素及其毒性进行了简单介绍;接着重点对微芯片技术在真菌毒素检测中的应用进行了详细论述;最后对微芯片应用于真菌毒素检测的发展前景和挑战进行了展望。     

Advanced synchrotron-based and globar-sourced molecular (micro) spectroscopy contributions to advances in food and feed research on molecular structure,mycotoxin determination,and molecular nutrition

ABSTRACT

Mycotoxin contamination has been a worldwide problem for food and feeds production for a long time. There is an obviously increased focus of the food and feed industry toward the reduction of mycotoxin concentration in both the raw materials and finished products. Therefore, both effective qualitative and quantitative techniques for the determination of mycotoxins are required to minimize their harmful effects. Conventional wet chemical methods usually are time-consuming, expensive, and rely on complex extraction and cleanup pretreatments. Synchrotron-based and globar-based molecular spectroscopy have shown great potential to be developed as rapid and nondestructive tools for the determination of molecular structure, molecular nutrition and mycotoxins in feed and food. This article reviews the common types of mycotoxins in feed and food, their toxicity, as well as the conventional detection methods. The principle of advanced molecular spectroscopy techniques and their application prospects for mycotoxin detection are discussed. Recent progress in food and feed research with molecular spectroscopy techniques is highlighted. This review provides a potential and insight into how to determine the structure and mycotoxins of feed and food on a molecular basis with advanced Synchrotron-based and globar-based molecular (micro) spectroscopy.     

Fungal and bacterial metabolites in commercial poultry feed from Nigeria

Metabolites of toxigenic fungi and bacteria occur as natural contaminants (e.g. mycotoxins) in feedstuffs making them unsafe to animals. The multi-toxin profiles in 58 commercial poultry feed samples collected from 19 districts in 17 states of Nigeria were determined by LC/ESI–MS/MS with a single extraction step and no clean-up. Sixty-three (56 fungal and seven bacterial) metabolites were detected with concentrations ranging up to 10,200?µg?kg^-1 in the case of aurofusarin. Fusarium toxins were the most prevalent group of fungal metabolites, whereas valinomycin occurred in more than 50% of the samples. Twelve non-regulatory fungal and seven bacterial metabolites detected and quantified in this study have never been reported previously in naturally contaminated stored grains or finished feed. Among the regulatory toxins in poultry feed, aflatoxin concentrations in 62% of samples were above 20?µg?kg^?1, demonstrating high prevalence of unsafe levels of aflatoxins in Nigeria. Deoxynivalenol concentrations exceeded 1000?µg?kg^?1 in 10.3% of samples. Actions are required to reduce the consequences from regulatory mycotoxins and understand the risks of the single or co-occurrence of non-regulatory metabolites for the benefit of the poultry industry.     

Natural Occurrence,Analysis, and Prevention of Mycotoxins in Fruits and their Processed Products

Mycotoxins are small toxic chemical products formed as the secondary metabolites by fungi that readily contaminate foods with toxins in the field or after harvest. The presence of mycotoxins, such as aflatoxins, ochratoxin A, and patulin, in fruits and their processed products is of high concern for human health due to their properties to induce severe acute and chronic toxicity at low-dose levels. Currently, a broad range of detection techniques used for practical analysis and detection of a wide spectrum of mycotoxins are available. Many analytical methods have been developed for the determination of each group of these mycotoxins in different food matrices, but new methods are still required to achieve higher sensitivity and address other challenges that are posed by these mycotoxins. Effective technologies are needed to reduce or even eliminate the presence of the mycotoxins in fruits and their processed products. Preventive measures aimed at the inhibition of mycotoxin formation in fruits and their processed products are the most effective approach. Detoxification of mycotoxins by different physical, chemical, and biological methods are less effective and sometimes restricted because of concerns of safety, possible losses in nutritional quality of the treated commodities and cost implications. This article reviewed the available information on the major mycotoxins found in foods and feeds, with an emphasis of fruits and their processed products, and the analytical methods used for their determination. Based on the current knowledge, the major strategies to prevent or even eliminate the presence of the mycotoxins in fruits and their processed products were proposed.     

农产品中真菌毒素检测方法研究进展

真菌毒素是由特定真菌在适宜条件下产生的有毒小分子次级代谢产物,农产品在生长、加工和储运过程中均存在被真菌毒素污染的可能性,被真菌毒素污染的农产品会严重危害动物和人类的健康。因此建立精确、高效的真菌毒素检测方法对于农产品中真菌毒素的防控和监测具有重要意义。本文主要介绍了真菌毒素的产生和危害,综述了近5年农产品中真菌毒素检测方法(仪器分析法、免疫学分析方法和光谱分析法)的研究进展,分析了这些检测方法的优缺点,并展望了农产品中真菌毒素在未来的发展趋势,为农产品中真菌毒素检测的相关研究和安全监管提供参考和启发。     

Specific antigen-based and emerging detection technologies of mycotoxins

Mycotoxins are secondary fungal metabolites produced by certain types of filamentous fungi or molds, such as Aspergillus, Fusarium, Penicillium, and Alternaria spp. Mycotoxins are natural contaminants of agricultural commodities, and their prevalence may increase due to global warming. According to the Food and Agriculture Organization of the United Nations, approximately 25% of the world's food crops are annually contaminated with mycotoxins. Mycotoxin-contaminated food and feed pose a high risk to both human and animal health. For instance, they possess carcinogenic, immunosuppressive, hepatotoxic, nephrotoxic, and neurotoxic effects. Hence, various approaches have been used to assess and control mycotoxin contamination. Significant challenges still exist because of the complex heterogeneous nature of food and feed composition. The potential of antigen-based approaches, such as enzyme-linked immunosorbent assay, flow injection immunoassay, chemiluminescence immunoassay, lateral flow immunoassay, and flow-through immunoassay, would contribute to our understanding about mycotoxins' rapid identification, their isolation, and the basic principles of the detection technologies. Additionally, we address other emerging technologies of potential application in the detection of mycotoxins. The data included in this review focus on basic principles and results of the detection technologies and would be useful as benchmark information for future research. © 2019 Society of Chemical Industry     

Fumonisins and fungi in dry soybeans (Glycine Max L.) for human consumption

This survey reports the occurrence of fumonisins (FBs) and fungi in dry soybeans sold for human consumption. The variation levels were 138–1495 µg kg^?1 and 178–552 µg kg^?1 for FB₁ and FB₂, respectively. In addition, potentially toxigenic fungi as Fusarium, Aspergillus and Penicillium genera were isolated in the samples. These can be considered as indicator-toxin and can produce considerable amounts of mycotoxins. Despite FB presence in the soybeans for human consumption, there is no legal regulation. Therefore, it is important to emphasise the need for frequent monitoring of these contaminants in soybeans.