Bio-inspired Sensor Fusion for Quality Assessment of Harumanis Mangoes

In recent years, there have been a number of reported studies on the use of non-destructive technique to evaluate and determine mango maturity and ripeness levels. However, most of these reported works were conducted using single-modality sensing systems, either using an electronic nose (e-nose), acoustics, CCD, IR sensor or by other non-destructive measurements. This paper presents the work on the classification of mangoes (Magnifera Indica cv. Harumanis) maturity and ripeness levels using data fusion of the electronic nose (e-nose) and acoustic sensor and combine with CCD and IR sensor. A Fourier-based shape separation method was developed from CCD camera images to grade mango by its shape and able to correctly classify 100%. Colour intensity from infrared image was used to distinguish and classify the level of maturity and ripeness of the fruits. The finding shows 92% correct classification of maturity levels by using infrared vision Three groups of samples each from two different harvesting times (week 7 and week 8) were evaluated by the e-nose and then followed by the acoustic sensor. By applying a low level data fusion technique on the e-nose and acoustic data, the classification for maturity and ripeness levels using LDA was improved.     

A comparison of different chemometrics approaches for the robust classification of electronic nose data

Accurate detection of certain chemical vapours is important, as these may be diagnostic for the presence of weapons, drugs of misuse or disease. In order to achieve this, chemical sensors could be deployed remotely. However, the readout from such sensors is a multivariate pattern, and this needs to be interpreted robustly using powerful supervised learning methods. Therefore, in this study, we compared the classification accuracy of four pattern recognition algorithms which include linear discriminant analysis (LDA), partial least squares-discriminant analysis (PLS-DA), random forests (RF) and support vector machines (SVM) which employed four different kernels. For this purpose, we have used electronic nose (e-nose) sensor data (Wedge et al., Sensors Actuators B Chem 143:365–372, 2009). In order to allow direct comparison between our four different algorithms, we employed two model validation procedures based on either 10-fold cross-validation or bootstrapping. The results show that LDA (91.56 % accuracy) and SVM with a polynomial kernel (91.66 % accuracy) were very effective at analysing these e-nose data. These two models gave superior prediction accuracy, sensitivity and specificity in comparison to the other techniques employed. With respect to the e-nose sensor data studied here, our findings recommend that SVM with a polynomial kernel should be favoured as a classification method over the other statistical models that we assessed. SVM with non-linear kernels have the advantage that they can be used for classifying non-linear as well as linear mapping from analytical data space to multi-group classifications and would thus be a suitable algorithm for the analysis of most e-nose sensor data.     

Analysis of colorectal adenocarcinoma tissue by desorption electrospray ionization mass spectrometric imaging

Negative ion desorption electrospray ionization (DESI) was used for the analysis of an ex vivo tissue sample set comprising primary colorectal adenocarcinoma samples and colorectal adenocarcinoma liver metastasis samples. Frozen sections (12 μm thick) were analyzed by means of DESI imaging mass spectrometry (IMS) with spatial resolution of 100 μm using a computer-controlled DESI imaging stage mounted on a high resolution Orbitrap mass spectrometer. DESI-IMS data were found to predominantly feature complex lipids, including phosphatidyl-inositols, phophatidyl-ethanolamines, phosphatidyl-serines, phosphatidyl-ethanolamine plasmalogens, phosphatidic acids, phosphatidyl-glycerols, ceramides, sphingolipids, and sulfatides among others. Molecular constituents were identified based on their exact mass and MS/MS fragmentation spectra. An identified set of molecules was found to be in good agreement with previously reported DESI imaging data. Different histological tissue types were found to yield characteristic mass spectrometric data in each individual section. Histological features were identified by comparison to hematoxylin-eosin stained neighboring sections. Ions specific to certain histological tissue types (connective tissue, smooth muscle, healthy mucosa, healthy liver parenchyma, and adenocarcinoma) were identified by semi-automated screening of data. While each section featured a number of tissue-specific species, no potential global biomarker was found in the full sample set for any of the tissue types. As an alternative approach, data were analyzed by principal component analysis (PCA) and linear discriminant analysis (LDA) which resulted in efficient separation of data points based on their histological types. A pixel-by-pixel tissue identification method was developed, featuring the PCA/LDA analysis of authentic data set, and localization of unknowns in the resulting 60D, histologically assigned LDA space. Novel approach was found to yield results which are in 95% agreement with the results of classical histology. KRAS mutation status was determined for each sample by standard molecular biology methods and a similar PCA/LDA approach was developed to assess the feasibility of the determination of this important parameter using solely DESI imaging data. Results showed that the mutant and wild-type samples fully separated. DESI-MS and molecular biology results were in agreement in 90% of the cases.     

钛铁矿型六方相ZnTiO3的电子结构和光学性质

分别采用基于密度泛函理论(DFT)的局域密度近似(LDA)和广义梯度近似(GGA)方法对钛铁矿型六方相ZnTiO₃的电子结构进行了第一性原理计算, 并在局域密度近似下计算了六方相ZnTiO₃的光学性质, 并将计算结果与实验数据进行了对比. 结果表明, 在局域密度近似下计算得到的结构参数更接近实验数据. 理论预测六方相ZnTiO₃属于直接带隙半导体材料, 其禁带宽度(布里渊区Z 点)为3.11 eV. 电子态密度和Mulliken 电荷布居分析表明Zn―O键是典型的离子键而Ti―O键是类似于钙钛矿型ATiO₃ (A=Sr, Pb, Ba)的Ti―O共价键. 在50 eV的能量范围内研究了ZnTiO₃的介电函数、吸收光谱和折射率等光学性质, 并基于电子能带结构和态密度对光学性质进行了解释.     

基于液体阵列味觉仿生传感器鉴别白酒香型的新方法

通过模拟哺乳动物的味觉系统, 建立了交叉响应的液体阵列传感器, 为鉴别白酒香型提供了新方法. 选用7种染料和1种卟啉化合物作为传感单元, 构建液体阵列传感器, 集合8个传感单元的光谱响应信号构成分析物的指纹图谱, 达到识别的目的. 使用96孔板酶标仪采集响应数据, 结合主成分分析(PCA)、分层聚类分析(HCA)和判别分析(LDA)等模式识别方法进行数据处理, 对9种具有代表性的不同香型白酒样品进行了鉴别分析. PCA结果表明, 该方法对于白酒的检测主要基于酒体微量成分, 其中酸类物质对识别的贡献最大(贡献率达54.3%), 芳香类物质贡献率为18.6%; 同时, 仅用63.4%的数据信息量即可对白酒香型进行区分. HCA结果表明, 平行样均正确归类, 各白酒之间的相似程度在聚类图上得到体现. LDA结果表明, 该阵列对于9种白酒样品香型识别的准确率达到100%.     

A Rapid and Efficient Way to Dynamic Creation of Cross‐Reactive Sensor Arrays Based on Ionic Liquids

Based on the simple counterion exchange of ionic liquids, a rapid, facile, and efficient strategy to create a cross‐reactive sensor array with a dynamic tunable feature was developed, and exemplified by the construction of a sensor array for the identification and classification of nitroaromatics and explosives mimics. To achieve a good sensing system with fast response, good sensitivity, and low detection limit, the synthesized ionic liquid receptors were tethered onto a silica matrix with a macro‐mesoporous hierarchical structure. Through the facile anion exchange approach, abundant ionic‐liquid‐based individual receptors with diversiform properties, such as different micro‐environments, diverse molecular interactions, and distinctive physico‐chemical properties, were easily and quickly synthesized to generate a distinct fingerprint of explosives for pattern recognition. The reversible anion exchange ability further endowed the sensor array with a dynamic tunable feature as well as good controllability and practicality for real‐world application. With the assistance of statistical analysis, such as principal component analysis (PCA) and linear discrimination analysis (LDA), an optimized‐size array with a good resolution was rationally established from a large number of IL‐based receptors. The performed experiments suggested that the ionic‐liquid‐based sensing protocol is a general and powerful strategy for creating a cross‐reactive sensor array that could find a wide range of applications for sensing various analytes or complex mixtures.     

Electronic properties of 3d transition metal dihalide monolayers predicted by DFT methods: Is there a pattern or are the results random?

We use periodic DFT calculations at LDA and PBE level to investigate 3d transition metal dihalide (TMDH) monolayers in H- and T-phase. By analyzing the phonon dispersion, we have obtained a rough overview which combinations may form stable structures. We have focused on identifying and explaining trends in the predicted electronic properties. Although their geometric structures are simple, the associated electronic and magnetic properties are not as easy to understand. At first glance, it seems that there is no clear trend, as even isovalence-electronic TMDH monolayers formed from the same metal but different halides can feature different magnetic moments. The identification of potential trends is further complicated by the fact that for a significant number of species, LDA results and PBE results predict different ground-state electronic structures. By rigorously analyzing the potential energy surfaces associated with different magnetic moments, we could show that the apparent inconsistencies can be easily understood as a result of the differences in the relative energy between electronic states of different magnetic moments. We further show that the trends in the band gaps can be easily rationalized by an electron counting rule based on simple symmetry arguments.     

A Novel Thin‐Film Copper Array Based on an Organic/Inorganic Sensor Hybrid: Microfabrication,Potentiometric Characterization,and Flow‐Injection Analysis Application

A first step towards the microfabrication of a thin‐film array based on an organic/inorganic sensor hybrid has been realized. The inorganic microsensor part incorporates a sensor membrane based on a chalcogenide glass material (Cu‐Ag‐As‐Se) prepared by pulsed laser deposition technique (PLD) combined with an PVC organic membrane‐based organic microsensor part that includes an o‐xylyene bis(N,N‐diisobutyl‐dithiocarbamate) ionophore. Both types of materials have been electrochemically evaluated as sensing materials for copper(II) ions. The integrated hybrid sensor array based on these sensing materials provides a linear Nernstian response covering the range 1×10^?6–1×10^?1 mol L^?1 of copper(II) ion concentration with a fast, reliable and reproducible response. The merit offered by the new type of thin‐film hybrid array includes the high selectivity feature of the organic membrane‐based thin‐film microsensor part in addition to the high stability of the inorganic thin‐film microsensor part. Moreover, the thin‐film sensor hybrid has been successfully applied in flow‐injection analysis (FIA) for the determination of copper(II) ions using a miniaturized home‐made flow‐through cell. Realization of the organic/inorganic thin‐film sensor hybrid array facilitates the development of a promising sophisticated electronic tongue for recognition and classification of various liquid media.     

Polymeric membrane ion-selective and cross-sensitive electrode-based electronic tongue for qualitative analysis of beverages

An electronic tongue based on the sensor array of polymeric membrane ion-selective electrodes combined with pattern recognition tools was applied to qualitative analysis of various brands of orange juice, tonic, and milk. The capability of this device to reliably discriminate between different brands of those products was presented. The tests of the system were performed using products of the same brand, but with different manufacture dates (and thus comparable by the term of taste). The fusion of two types of sensors-classical selective ones and partially selective in one versatile array, and working out the sensor array's response by means of principal component analysis and back propagation neural network methods allowed the discrimination between different brands of various beverages with very high accuracy (90-100%). The real performance of the electronic tongue was evaluated applying testing samples from another manufacture lot, than the samples used in the learning set.     

Characterization and classification of the aroma of beer samples by means of an MS e-nose and chemometric tools

An electronic nose based on coupling of headspace (HS) with a mass spectrometer (MS) has been used in this study to classify and characterize a series of beers according to their production site and chemical composition. With this objective, we analyzed 67 beers of the same brand and preparation process but produced in different factories. The samples were also subjected to sensory evaluation by a panel of experts. Linear discriminant analysis (LDA) was used as the classification technique and stepwise LDA based on Wilk’s lambda criterion was used to select the most discriminating variables. To interpret the aroma characteristics of the beers from the m/z ions obtained, score and loading bi-plots were obtained by applying canonical variables. Because the beers analyzed were marketed with the same name and brand, we expected to be working with the same product irrespective of its origin. However, results from both sensory evaluation and use of the e-nose revealed differences between factories. With the e-nose it was possible to relate these differences to the presence (and abundance) of characteristic ions of different compounds typically found in beer. These results demonstrate that the HS–MS e-nose is not only an aroma sensor capable to classify and/or differentiate samples but it can also provide information about the compounds responsible for this differentiation.     

Comparison of methods for outlier detection and their effects on the classification results for a particular data base

One of the drawbacks for using linear discriminant analysis (LDA) is the presence of outliers. Some methods of detecting outliers are compared and applied to a particular data base. When multivariate methods (multinormal distribution procedure and Hawkins' procedure) were applied, the two subsets produced did not differ greatly. Assumptions needed for the application of LDA were evaluated for each subset. Classification ability, feature selection and prediction ability were considered for each subset. Results for each subset were quite different. Hawkins' procedure seems the better method for detecting outliers.     

Aun (n＝2～20)团簇的遗传算法和密度泛函方法研究

在遗传算法和Gupta多体势系统地搜索金属团簇初始结构基础上, 应用密度泛函理论和基于局域密度近似 (LDA)或广义梯度近似(GGA)的超软赝势和投影扩充波(PAW)方法分别系统地研究了金属团簇Au_n (n≤20)的最稳定构型和电子性质. 发现LDA或GGA近似下, 最稳定构型存在一定的差异: LDA方法中, Au团簇最稳定构型从Au₇ 处就发生了从二维结构向三维结构的转化, 而GGA近似下Au₁₃的最稳定异构体仍然保持平面构型. 计算结果表明, 平均最近邻距、平均配位数和结合能随着尺寸的增大呈递增趋势, 而二阶差分能、费米能级、HOMO-LUMO能隙、垂直电离势和电子亲和势出现了明显的奇偶交替现象. 其结果丰富了目前对金团簇的理论和实验的研究.     

Revealing the properties of oils from their dissolved hydrocarbon compounds in water with an integrated sensor array system

This paper presents a system and method developed to identify a source oil's characteristic properties by testing the oil's dissolved components in water. Through close examination of the oil dissolution process in water, we hypothesise that when oil is in contact with water, the resulting oil-water extract, a complex hydrocarbon mixture, carries the signature property information of the parent oil. If the dominating differences in compositions between such extracts of different oils can be identified, this information could guide the selection of various sensors, capable of capturing such chemical variations. When used as an array, such a sensor system can be used to determine parent oil information from the oil-water extract. To test this hypothesis, 22 oils' water extracts were prepared and selected dominant hydrocarbons analyzed with Gas Chromatography-Mass Spectrometry (GC-MS); the subsequent Principal Component Analysis (PCA) indicates that the major difference between the extract solutions is the relative concentration between the volatile mono-aromatics and fluorescent polyaromatics. An integrated sensor array system that is composed of 3 volatile hydrocarbon sensors and 2 polyaromatic hydrocarbon sensors was built accordingly to capture the major and subtle differences of these extracts. It was tested by exposure to a total of 110 water extract solutions diluted from the 22 extracts. The sensor response data collected from the testing were processed with two multivariate analysis tools to reveal information retained in the response patterns of the arrayed sensors: by conducting PCA, we were able to demonstrate the ability to qualitatively identify and distinguish different oil samples from their sensor array response patterns. When a supervised PCA, Linear Discriminate Analysis (LDA), was applied, even quantitative classification can be achieved: the multivariate model generated from the LDA achieved 89.7% of successful classification of the type of the oil samples. By grouping the samples based on the level of viscosity and density we were able to reveal the correlation between the oil extracts' sensor array responses and their original oils' feature properties. The equipment and method developed in this study have promising potential to be readily applied in field studies and marine surveys for oil exploration or oil spill monitoring.     

Fluorometric sensing of biogenic amines with aggregation-induced emission-active tetraphenylethenes

A fluorometric sensor for detection and identification of biogenic amines with carboxylic acid modified tetraphenylethenes (TPEs) based on aggregation-induced emission (AIE) is reported. A mixture of the carboxylic acid substituted TPE and biogenic amines displayed a blue emission on aggregation, which serves as a "turn-on" fluorescent sensor for the amines, the degree of fluorescence enhancement being dependent on the amine. The chromic responses were utilized to distinguish the amines. A fluorometric sensor array of three TPEs with carboxylic acid groups was shown to identify accurately 10 different amines, including biogenic amines. The response patterns were systematically classified by using linear discriminant analysis (LDA) with 98% classification accuracy. Additional information on the concentration of histamine in a "tuna fish matrix" as an example was assessed by the further analysis of the fluorescence intensity, demonstrating a test for food freshness and quality.     

Ensemble of linear models for predicting drug properties

We propose a new classification method for the prediction of drug properties, called random feature subset boosting for linear discriminant analysis (LDA). The main novelty of this method is the ability to overcome the problems with constructing ensembles of linear discriminant models based on generalized eigenvectors of covariance matrices. Such linear models are popular in building classification-based structure-activity relationships. The introduction of ensembles of LDA models allows for an analysis of more complex problems than by using single LDA, for example, those involving multiple mechanisms of action. Using four data sets, we show experimentally that the method is competitive with other recently studied chemoinformatic methods, including support vector machines and models based on decision trees. We present an easy scheme for interpreting the model despite its apparent sophistication. We also outline theoretical evidence as to why, contrary to the conventional AdaBoost ensemble algorithm, this method is able to increase the accuracy of LDA models.     

Discrimination of coatings on wooden materials using the gas sensor system

The applicability of sensor system for the discrimination of sources of indoor pollution was investigated. As examples of indoor pollution sources, paint and lacquer coatings were considered. Commercially available preparations: Akrylux, Doamlux, Bejca and White Scandinavian were selected for headspace measurements using TGS sensor array. Following issues were investigated: (1) discrimination between water- and solvent-based coatings, (2) discrimination between one component coatings, and (3) discrimination between one component and two component coatings. Following data analysis methods were used: principal component analysis (PCA), linear discriminant analysis (LDA) and probabilistic neural network (PNN). Results showed that coatings could be discriminated successfully, provided the surface covered was solid wood (0-1.8% error). The interference of fibreboard volatiles in sensor measurements of coatings was most likely encountered. It could have significantly impaired discrimination of coatings on fibreboard (2.8-5.6% error) as compared to wood. Worst results were obtained for the discrimination of coatings on unknown material(12.5-28.7% error).     

VOCs分子的半导体型传感器识别检测研究进展

具有体积小、功耗低、灵敏度高、硅工艺兼容性好等优点的金属氧化物半导体(MOS)气体传感器现已广泛地应用于军事、科研和国民经济的各个领域。然而MOS传感器的低选择性阻碍了其在物联网(IoT)时代的应用前景。为此,本文综述了解决MOS传感器选择性的研究进展,主要介绍了敏感材料性能提升、电子鼻和热调制三种改善MOS传感器选择性的技术方法,阐述了三种方法目前所存在的问题及其未来的发展趋势。同时,本文还对比介绍了机器嗅觉领域主流的主成分分析(PCA)、线性判别分析(LDA)和神经网络(NN)模式识别/机器学习算法。最后,本综述展望了具有数据降维、特征提取和鲁棒性识别分类性能的卷积神经网络(CNN)深度学习算法在气体识别领域的应用前景。基于敏感材料性能的提升、多种调制手段与阵列技术的结合以及人工智能(AI)领域深度学习算法的最新进展,将会极大地增强非选择性MOS传感器的挥发性有机化合物(VOCs)分子识别能力。     

Gas sensor array based on metal-decorated carbon nanotubes

Here we demonstrate design, fabrication, and testing of electronic sensor array based on single-walled carbon nanotubes (SWNTs). Multiple sensor elements consisting of isolated networks of SWNTs were integrated into Si chips by chemical vapor deposition (CVD) and photolithography processes. For chemical selectivity, SWNTs were decorated with metal nanoparticles. The differences in catalytic activity of 18 catalytic metals for detection of H(2), CH(4), CO, and H(2)S gases were observed. Furthermore, a sensor array was fabricated by site-selective electroplating of Pd, Pt, Rh, and Au metals on isolated SWNT networks located on a single chip. The resulting electronic sensor array, which was comprised of several functional SWNT network sensors, was exposed to a randomized series of toxic/combustible gases. Electronic responses of all sensor elements were recorded and the sensor array data was analyzed using pattern-recognition analysis tools. Applications of these small-size, low-power, electronic sensor arrays are in the detection and identification of toxic/combustible gases for personal safety and air pollution monitoring.     

A single cataluminescence sensor based on spectral array and its use in the identification of vinegars

The discrimination of complex mixtures, especially those with very similar compositions, remains a challenging part of chemical analysis. In this paper, a single cataluminescence (CTL) sensor constructed using MgO nanomaterials in a closed reaction cell (CRC) was used to identify vinegars. It may provide an archetype of this type of highly multicomponent system. By scanning the CTL spectra, which were distributed in 15 wavelengths during the reaction period, the spectral array patterns of the vinegars were obtained. These functioned as their fingerprints. The CTL signals of the array were then normalized and identified through linear discrimination analysis (LDA). Nine types and eight brands of vinegars and an additional series of artificial samples were tested; the new technique was found to distinguish between them very well. This single sensor demonstrated excellent promise for analysis of complex mixtures in real-world applications and may provide a novel method for identifying very similar complex analytes.