Application of artificial neural network in food classification

Artificial neural network (ANN) classifiers have been successfully implemented for various quality inspection and grading tasks of diverse food products. ANN are very good pattern classifiers because of their ability to learn patterns that are not linearly separable and concepts dealing with uncertainty, noise and random events. In this research, the ANN was used to build the classification model based on the relevant features of beer. Samples of the same brand of beer but with varying manufacturing dates, originating from miscellaneous manufacturing lots, have been represented in the multidimensional space by data vectors, which was an assembly of 12 features (% of alcohol, pH, % of CO(2) etc.). The classification has been performed for two subsets, the first that included samples of good quality beer and the other containing samples of unsatisfactory quality. ANN techniques allowed the discrimination between qualities of beer samples with up to 100% of correct classifications.     

Supervised pattern recognition in food analysis

Data analysis has become a fundamental task in analytical chemistry due to the great quantity of analytical information provided by modern analytical instruments. Supervised pattern recognition aims to establish a classification model based on experimental data in order to assign unknown samples to a previously defined sample class based on its pattern of measured features. The basis of the supervised pattern recognition techniques mostly used in food analysis are reviewed, making special emphasis on the practical requirements of the measured data and discussing common misconceptions and errors that might arise. Applications of supervised pattern recognition in the field of food chemistry appearing in bibliography in the last two years are also reviewed.     

Pattern recognition based on fuzzy observations for spectroscopic quality control and chromatographic fingerprinting

The quality of multicomponent samples from one or several groups of samples can be monitored by a pattern recognition method. The method is based on profiles of sample quality, which are obtained by means of a multicomponent analytical technique (e.g., ultraviolet spectroscopy or chromatography), and data reduction is done with the aid of fuzzy set theory. The advantages of the method in cases of overlapping and non-additive signals are outlined for quality control of analgesic tablets by ultraviolet spectroscopy. Its performance in the case of highly uncertain data patterns is demonstrated for classification of protein samples by chromatography.     

Characterization of Galician (N.W. Spain) quality brand potatoes: a comparison study of several pattern recognition techniques

Authenticity is an important food quality criterion and rapid methods to guarantee it are widely demanded by food producers, processors, consumers and regulatory bodies. The objective of this work was to develop a classification system in order to confirm the authenticity of Galician potatoes with a Certified Brand of Origin and Quality (CBOQ) 'Denominación Específica: Patata de Galicia' and to differentiate them from other potatoes that did not have this CBOQ. Ten selected metals were determined by atomic spectroscopy in 102 potato samples which were divided into two categories: CBOQ and non-CBOQ potatoes. Multivariate chemometric techniques, such as cluster analysis and principal component analysis, were applied to perform a preliminary study of the data structure. Four supervised pattern recognition procedures [including linear discriminant analysis (LDA), K-nearest neighbours (KNN), soft independent modelling of class analogy (SIMCA) and multilayer feed-forward neural networks (MLF-ANN)] were used to classify samples into the two categories considered on the basis of the chemical data. Results for LDA, KNN and MLF-ANN are acceptable for the non-CBOQ class, whereas SIMCA showed better recognition and prediction abilities for the CBOQ class. A more sophisticated neural network approach performed by the combination of the self-organizing with adaptive neighbourhood network (SOAN) and MLF network was employed to optimize the classification. Using this combined method, excellent performance in terms of classification and prediction abilities was obtained for the two categories with a success rate ranging from 98 to 100%. The metal profiles provided sufficient information to enable classification rules to be developed for identifying potatoes according to their origin brand based on SOAN-MLF neural networks.     

Partial least-squares regression and fuzzy clustering — A joint approach

Chemical and physical analyses of malt, the main ingredient of beer, have been used to predict the concentration of certain volatile compounds in the finished beer.The prediction was done by means of the partial least squares regression (PLS2) in SIMCA. The total data set as well as individual malt clusters were submitted to PLS analysis. Best prediction was obtained by separating the total object matrix in classes according to similarity found by fuzzy pattern recognition (FCV). FCV was also used to separate the beer variables in classes and to select the subset of variables to be predicted.A joint approach of fuzzy pattern recognition to identify groups of samples and SIMCA-PLS2 to predict several dependent variables is suggested as a powerful tool in process-analytical chemistry.     

Rapid identification of the quality decoction pieces by partial least squares ‐based pattern recognition: grade classification of the decoction pieces of Saposhnikovia divaricata

Herbal medicines are commonly used in many countries after they undergo processing. Quality decoction pieces are a guarantee of the efficacy and safety of the herbal medical products. Here, a strategy based on chemical analysis combined with chemometric techniques was proposed for the classification and prediction of the different grades of the decoction pieces. Considering the necessity for a shared and simple method for the grade classification for the public, in this paper, the characterization of the chemical constituents was determined by utilizing high‐performance liquid chromatography (HPLC)/diode array detection. HPLC was first established for the characterization of the chemical constituents of the different grade decoction pieces. Furthermore, a simultaneous quantification of several of the marker compounds in these decoction pieces was obtained. Finally, a partial least squares‐based pattern recognition method was utilized to obtain a predictive model for the grade classification of the decoction pieces. Saposhnikovia divaricata (Turcz.) Schischk was used as a case study. The partial least squares ‐based pattern recognition for the grade classification of the decoction pieces of S. divaricata demonstrated good sensitivity, specificity and prediction performance, which may efficiently validate the identification results of appearance assessment. The proposed strategy is expected to provide a new insight for the grade classification and quality control of the decoction pieces. Copyright © 2016 John Wiley & Sons, Ltd.     

模式识别在食品质量控制方面的应用进展

本文介绍了食品质量研究中常用的一些化学模式识别方法的基本原理,并介绍了模式识别结合红外、原子吸收、原子发射、气相色谱、液相色谱、质谱、电子鼻传感器等检测技术在食品质量控制中的应用.对化学计量学在食品质量控制中的应用前景作了展望.     

Nutritional value of trace elements in spaghetti sauces and their classification according to the labeled taste using pattern recognition techniques

Summary The nutrient trace elements chromium, iron and zinc as well as cobalt, rubidium and scandium were determined in dry spaghetti sauce samples from the Greek market by instrumental neutron activation analysis. The results were evaluated according to the new US Recommended Dietary Allowances (RDA), US Adequate Intake (AI), US Reference Values for nutrition labeling (RVNL) and European Union reference values for nutrition labeling (EURV). Moreover, the same data has been used with pattern recognition techniques in order to classify the sauce samples according to their labeled flavor. The evaluation showed that the nutrition rate depends strongly on the reference value under consideration. The spaghetti sauces studied are a good source for the covering of chromium daily AI. The same sauces are poor source for zinc daily needs of the organism (RDA, RVNL), but they are a moderate source for iron daily needs (RDA). The application of cluster analysis, of linear discriminant analysis and of the principal component analysis classified the spaghetti sauce samples according to their labeled taste successfully. In addition using the same techniques, another classification in red and white spaghetti sauces is carried out according to their tomato content.     

Study of Chromatographic Fingerprint of the Flavor in Beer by HS-SPME-GC

The chromatographic fingerprint of the flavor in beer, which was obtained by analyzing 28 beer samples (6 tastes) from 4 breweries, was established by Headspace Solid-phase Microextraction coupled with Gas Chromatogram Flame Ionization Detector (HS-SPME-GC-FID). After Principal Component Analysis (PCA) and Hierarchical Clustering Analysis (HCA) were used to process the GC data, not only could 28 beer samples be classified into three main types (draft beer, traditional beer, and dark beer), but also the same main type of beer samples could be further classified individually according to the breweries and tastes. In addition, 18 volatile compounds were identified by Gas Chromatogram Mass Spectrum (GC-MS). The results showed that HS-SPME-GC-FID was convenient, rapid, and precise in classifying beer samples according to different main types, breweries, and even tastes. Therefore, the method established in this paper was potential to be used for the identification of beer types and even the quality control of beer.     

UV–化学模式识别不同产地的丹参

建立由UV–化学模式识别法评价丹参质量的方法。分别用正己烷、乙酸乙酯、水、乙醇提取不同产地的丹参,并测绘其紫外光谱。取紫外光谱各波长的吸光度为特征数据,进行主成分分析、聚类分析,对不同产地丹参质量的差异进行了评价。不同溶剂提取液的光谱聚类结果有所差异,可将不同产地丹参聚为3类。UV–化学模式识别技术可以从整体上反应丹参所含成分的差异,可为丹参质量控制与评价提供依据。     

Two-dimensional wavelet analysis based classification of gas chromatogram differential mobility spectrometry signals

This study introduces two-dimensional (2-D) wavelet analysis to the classification of gas chromatogram differential mobility spectrometry (GC/DMS) data which are composed of retention time, compensation voltage, and corresponding intensities. One reported method to process such large data sets is to convert 2-D signals to 1-D signals by summing intensities either across retention time or compensation voltage, but it can lose important signal information in one data dimension. A 2-D wavelet analysis approach keeps the 2-D structure of original signals, while significantly reducing data size. We applied this feature extraction method to 2-D GC/DMS signals measured from control and disordered fruit and then employed two typical classification algorithms to testify the effects of the resultant features on chemical pattern recognition. Yielding a 93.3% accuracy of separating data from control and disordered fruit samples, 2-D wavelet analysis not only proves its feasibility to extract feature from original 2-D signals but also shows its superiority over the conventional feature extraction methods including converting 2-D to 1-D and selecting distinguishable pixels from training set. Furthermore, this process does not require coupling with specific pattern recognition methods, which may help ensure wide applications of this method to 2-D spectrometry data.     

Characterization of winemaking yeast by cell number-size distribution analysis through flow field-flow fractionation with multi-wavelength turbidimetric detection

Yeasts are widely used in several areas of food industry, e.g. baking, beer brewing, and wine production. Interest in new analytical methods for quality control and characterization of yeast cells is thus increasing. The biophysical properties of yeast cells, among which cell size, are related to yeast cell capabilities to produce primary and secondary metabolites during the fermentation process. Biophysical properties of winemaking yeast strains can be screened by field-flow fractionation (FFF). In this work we present the use of flow FFF (FlFFF) with turbidimetric multi-wavelength detection for the number-size distribution analysis of different commercial winemaking yeast varieties. The use of a diode-array detector allows to apply to dispersed samples like yeast cells the recently developed method for number-size (or mass-size) analysis in flow-assisted separation techniques. Results for six commercial winemaking yeast strains are compared with data obtained by a standard method for cell sizing (Coulter counter). The method here proposed gives, at short analysis time, accurate information on the number of cells of a given size, and information on the total number of cells.     

Pattern recognition in analytical chemistry

Pattern recognition techniques are a group of modern mathematical methods for solving classification problems. In analytical chemistry they may well become essential tools for the automation of some types of classification problems, such as qualitative analysis by the interpretation of spectra, or quality control problems in, for instance, the food industry.     

Exemplar application of multi-capillary column ion mobility spectrometry for biological and medical purpose

In recent years, ion mobility spectrometry is increasingly in demand for new applications especially on biological samples (cells, bacteria, fungi), in medicine (diagnosis, therapy and medication control e.g. from breath analyses), for food quality control, safety monitoring and characterisation or process control in chemical and pharmaceutical industry. For this purpose instruments based on gas phase separation of ions in weak electric fields were developed at ISAS–Institute for Analytical Sciences, focussing on the particular challenges such as humid and rather complex samples, specific sampling procedures adapted to the application, fast pre-separation techniques like multi-capillary columns and suitable data processing including data bases for relevant analytes and automatic characterisation of IMS-chromatograms. Feasibility studies were carried out successfully for biological and medical purpose at ISAS, including the detection of bacteria, fungi and metabolites of cells and in human breath. For all those samples characteristic pattern of analytes were found and could be used for the identification of cell lines, fungi and bacteria as well as of numerous diseases. Furthermore, the quantification of those analytes could be used to obtain information about the state of the process or person (e.g. growth of cultures, development of diseases, level of medication, grade of cancer). Those examples shall demonstrate the potential of ion mobility spectrometry for the selected applications. However, a general and reliable data bases of reference analytes is required in the near future to enable an exploitation of the metabolic pathways and to confirm the relevance of the detected signals for the investigated topic.     

Fractionation analysis of manganese and zinc in beers by means of two sorbent column system and flame atomic absorption spectrometry

In the present article, a method of operational fractionation of Mn and Zn in beer using flame atomic absorption spectrometry was developed. The proposed fractionation scheme was based on use of a hydrophobic adsorbing resin Amberlite XAD7 (first column, 2 g resin bed) connected in a series with a strong cation exchanger Dowex 50Wx4 (second column, 1 g resin bed). After passing the samples of beers through the columns, distinct groupings of Mn and Zn species retained on the sorbents, i.e., hydrophobic fraction of polyphenols bound metal species and cationic metal species fraction, respectively, were determined in respective eluates obtained after complete recovery of Mn and Zn species with 10 ml of 2.0 mol l⁻¹ HNO₃ (first column) and 10 ml of 4.0 mol l⁻¹ HCl (second column). In addition, the effluents collected were analyzed prior to the evaluation of the third, residual fraction, presumably attributed to any hydrophilic anionic and inert metal species. The established fractionation patterns for Mn and Zn were discussed in reference to likely associations of metals with endogenous food bioligands and possible availability of the distinguished metal species classes. The quality of the results was proved by the recovery experiments.     

Data analysis for electronic nose systems

Electronic noses (e-noses) employ an array of chemical gas sensors and have been widely used for the analysis of volatile organic compounds. Pattern recognition provides a higher degree of selectivity and reversibility to the systems leading to an extensive range of applications. These range from the food and medical industry to environmental monitoring and process control. Many types of data analysis techniques have been used on the data produced. This review covers aspects of analysis from data normalisation methods to pattern recognition and classification techniques. An overview of data visualisation such as non-linear mapping and multivariate statistical techniques is given. Focus is then on the use of artificial intelligence techniques such as neural networks and fuzzy logic for classification and genetic algorithms for feature (sensor) selection. Application areas are covered with examples of the types of systems and analysis methods currently in use. Future trends in the analysis of sensor array data are discussed.     

基于主成分分析-Fisher判别分析的食品类塑料瓶物证差分拉曼光谱分类

食品类塑料瓶物证携带许多潜在证据信息,目前针对此类物证的检验研究尚处于探索阶段。利用差分拉曼光谱对46个食品类塑料瓶样品进行检验,依据样品材质及光谱特征峰可将样品分为三类。利用主成分分析（Principal component analysis, PCA）-Fisher判别分析,绘制主成分得分图,构建判别函数,建立分类模型。结果表明,食品类塑料瓶样品具有明显的聚类关系,原始分类与交叉验证分类准确率达到100 %。差分拉曼光谱结合PCA-Fisher判别分析,检验鉴别食品类塑料瓶物证具有一定的科学性。     

Chromatographic and electrophoretic approaches for the analysis of protein quality of soy beverages

Furosine, generated by acid hydrolysis of fructosyllysine, an early Maillard reaction product, is a highly valuable indicator of food quality and, more specifically, of food protein quality. Ion pair RP-HPLC and CZE techniques were employed to determine furosine content in beverages based on soymilk (n = 15) and cow's milk supplemented with soy isoflavones (n = 1). The levels of furosine found in the samples ranged from 25.55 +/- 0.18 to 170.72 +/- 10.4 mg/100 g of protein by HPLC and from 28.67 +/- 1.84 to 161.25 +/- 5.78 mg/100 g of protein by CZE. Results obtained by both analytical techniques do not differ significantly (p > 0.05), confirming their feasibility for furosine analysis in soy-based products.