Predicting Vodka Adulteration: A Combination of Electronic Tongue and Artificial Neural Networks

An artificial neural network was used to build models caple of predicting and quantifying vodka adulteration with methanol and/or tap water. A voltammetric electronic tongue based on gold and copper microelectrodes was used, and 310 analyses were performed. Vodkas were adulterated with tap water (5 to 50% (v/v)), methanol (1 to 13% (v/v)), and with a fixed addition of 5% methanol and tap water varying from 5 to 50% (v/v). The classification model showed 99.5% precision, and it correctly predicted the type of adulterant in all samples. Regarding the regression model, the root mean squared error was 3.464% and 0.535% for the water and methanol addition, respectively, and the prediction of the adulterant content presented an R2 0.9511 for methanol and 0.9831 for water adulteration.

Wearable Electronic Tongue for Non-Invasive Assessment of Human Sweat

Sweat is a promising biofluid in allowing for non-invasive sampling. Here, we investigate the use of a voltammetric electronic tongue, combining different metal electrodes, for the purpose of non-invasive sample assessment, specifically focusing on sweat. A wearable electronic tongue is presented by incorporating metal electrodes on a flexible circuit board and used to non-invasively monitor sweat on the body. The data obtained from the measurements were treated by multivariate data processing. Using principal component analysis to analyze the data collected by the wearable electronic tongue enabled differentiation of sweat samples of different chemical composition, and when combined with 1H-NMR sample differentiation could be attributed to changing analyte concentrations.

Locally Linear Embedding as Nonlinear Feature Extraction to Discriminate Liquids with a Cyclic Voltammetric Electronic Tongue

Electronic tongues are devices used in the analysis of aqueous matrices for classification or quantification tasks. These systems are composed of several sensors of different materials, a data acquisition unit, and a pattern recognition system. Voltammetric sensors have been used in electronic tongues using the cyclic voltammetry method. By using this method, each sensor yields a voltammogram that relates the response in current to the change in voltage applied to the working electrode. A great amount of data is obtained in the experimental procedure which allows handling the analysis as a pattern recognition application; however, the development of efficient machine-learning-based methodologies is still an open research interest topic. As a contribution, this work presents a novel data processing methodology to classify signals acquired by a cyclic voltammetric electronic tongue. This methodology is composed of several stages such as data normalization through group scaling method and a nonlinear feature extraction step with locally linear embedding (LLE) technique. The reduced-size feature vector input to a k-Nearest Neighbors (k-NN) supervised classifier algorithm. A leave-one-out cross-validation (LOOCV) procedure is performed to obtain the final classification accuracy. The methodology is validated with a data set of five different juices as liquid substances.Two screen-printed electrodes voltametric sensors were used in the electronic tongue. Specifically the materials of their working electrodes were platinum and graphite. The results reached an 80% classification accuracy after applying the developed methodology.

Simultaneous Sensing of Codeine and Diclofenac in Water Samples Using an Electrochemical Bi-MIP Sensor and a Voltammetric Electronic Tongue

Codeine and diclofenac overdoses have been widely reported. Here, a biomimetic sensor (bi-MIP) was devised, and an electronic tongue was used to analyze water samples simultaneously containing both these drugs. The bi-MIP sensor limits of detection for diclofenac and codeine taken individually were 0.01 µg/mL and 0.16 µg/mL, respectively. Due to a cross-reactivity effect when using the bi-MIP sensor, the electronic tongue was shown to differentiate samples containing both analytes. The results confirm the feasibility of simultaneous detection of two target analytes via a bi-MIP sensor. Additionally, they demonstrate the ability of a multi-sensor to classify different water samples.

Detection of Biogenic Amines in Canned Tuna Using a Voltammetric Electronic Tongue

Biogenic amines (BAs), which are produced naturally due to the decomposition of amino acids, are crucial for the food industry because its formation is directly related to improper storage and the presence of bacteria. High concentrations of BAs can be easily related with the quality and spoilage of the products of this sector. The necessity to quickly and efficiently quantify these targets makes mandatory the use of alternatives to conventional analytical methods used up to now. For example, the combination of sensors with chemometric tools (known as electronic tongue) are a promising alternative for quick and informative analysis in the food sector. Chemometric tools allow us to develop models for the quantification of specific compounds in a complex matrix, making it a feasible tool for the development of more user-friendly methods than the traditional ones. In this context, the work presents an electronic tongue created for the detection of histamine, cadaverine and tyramine using a set of five modified GEC (graphite epoxy composite) electrodes: ZnO NPs, CuO NPs, SnO2 NPs, Bi2O3 NPs, and polypyrrole, as the voltammetric multisensor array. The chemometric model was obtained with an Artificial Neural Network (ANN) with 51 input neurons, five neurons in the hidden layer and three neurons in the output layer. The functions used for the hidden and output layers were tansig and purelin, respectively. The results show slopes near to 1 and intercepts close to 0, indicating the feasibility of the model.