基于FTIR-CWT及ANN分类法的中药材菟丝子与同科属金灯藤子的识别方法研究

采用水平衰减全反射傅里叶变换红外光谱法(HATR-FTIR)测定了种子植物中药材菟丝子及同科属种子植物金灯藤子的FTIR,运用基于连续小波变换多分辨率分析法对吸收较为相似的菟丝子及金灯藤子的FTTR进行特征提取.通过分析比较后选择第7,10,13分解层的特征向量,进行人工神经网络(ANN)训练,再用训练出来的网络对不同产地的植物种子菟丝子和金灯藤子所得FTIR小波提取的特征向量进行分类.通过对32个不同样本的验证,说明能够采用基于FTIR-连续小波特征提取及人工神经网络分类法对同科属中药材菟丝子与金灯藤子进行识别.     

离散小波特征提取及人工神经网络分类法的傅里叶变换红外光谱法识别鳞毛蕨科3种植物

利用水平衰减全反射-傅里叶变换红外光谱法测定了3种药用鳞毛蕨科植物贯众、阔鳞鳞毛蕨和变异鳞毛蕨根部的FT-IR。运用基于离散小波多分辨率分析法对FT-IR吸收较为相似的3种药用蕨类植物根的FT-IR进行特征提取。选择第4、5分解层数的特征向量,进行人工神经网络(Artificial neural network,ANN)训练;再用训练出来的网络对不同产地的3种药用蕨类植物根所得FT-IR小波提取的特征向量进行分类。通过对240个不同样本的预测,说明能够采用基于FT-IR-离散小波特征提取及人工神经网络分类法对同科3种药用蕨类植物根的FT-IR进行识别。     

应用小波特征提取肺癌组织FTIR的支持向量机分类方法研究

提出了一种新的基于傅里叶变换红外光谱(Fourier Transform Infrared Spectroscopy, FTIR)的小波特征提取与支持向量机(SVM)分类方法以提高FTIR对早期肺癌的诊断准确率. 对肺正常组织、早期肺癌及进展期肺癌组织的FTIR, 利用连续小波(CW)多分辨率分析法提取9个特征量, 支持向量机把其分为正常组与非正常组(包括早期肺癌和进展期肺癌), 对正常组织、早期肺癌和进展期肺癌的识别, 多项式核函数和径向基函数的识别准确率最高. 多项式核函数对正常组织、早期肺癌和进展期肺癌的识别准确率分别为100%, 95%及100%; 径向基函数分别为100%, 95%和100%. 实验结果表明FTIR-CW-SVM模式分类方法对正常肺癌组织、早期肺癌及进展肺癌的识别具有较好的可行性.     

一种新颖的基于FTIR-CWT及ANN分类法的同科属中药材菟丝子与金灯藤的识别研究

采用水平衰减全反射傅里叶变换红外光谱法（HATR-FTIR）测定了同属种子植物中药材菟丝子及金灯藤的FTIR,运用基于连续小波多分辨率分析法对吸收较为相似的菟丝子及金灯藤的FTIR进行特征提取。选择第7、10、13分解层数的特征向量,进行人工神经网络（ANN）训练,再用训练出来的网络对不同产地的植物种子菟丝子和金灯藤所得FTIR小波提取的特征向量进行分类。通过对32个不同样本的验证,说明能够采用基于FTIR-连续小波特征提取及人工神经网络分类法对同科属中药材菟丝子与金灯藤进行识别。     

烟草组分的近红外光谱和支持向量机分析

测定了120个产自福建、安徽和云南烟草样品的近红外光谱. 在利用支持向量机(SVM)技术建立其定量、定性分析模型之前, 用小波变换技术对光谱变量进行了有效的压缩, 然后采用径向基核函数建立了75个烟草样品的分类模型, 同时建立了总糖、还原糖、烟碱和总氮4个组分的定量分析模型, 并利用45个烟草样品对模型进行了检验. 仿真实验表明, 建立的SVM分类模型分类准确率达到100%, 而4个组分的定量分析模型的预测决定系数(R2)、预测均方差(RMSEP)和平均相对误差(RME)3个指标值显示其模型泛化能力非常强, 预测效果良好, 可见这是一种有效的近红外光谱的建模分析方法.     

Classification of Rat FTIR Colon Cancer Data Using Waveletsand BPNN

本文提出了一种新的基于水平衰减全反射-傅里叶变换红外光谱（HATR-FTIR）的小波特征提取与反向传播人工神经网络模式分类方法以提高FTIR对早期大鼠结肠癌的诊断准确率．对60只DMH诱导的SD大鼠,44只诱导鼠的第二代鼠,36只正常SD大鼠的结肠正常组织、异常增生、早癌及进展期癌组织所获得的的HATR-FTIR,利用连续小波多尺度分析法提取12个特征量,采用反向传播人工神经网络进行分类,识别准确率分别为100%、94%、97.5%及100%．实验结果表明此方法对早期结肠癌具有较高的诊断率．     

水平衰减全反射-傅里叶变换红外光谱结合化学计量学法应用于三种缩叶藓属植物的分类

采用水平衰减全反射(HATR)-傅里叶变换红外光谱法(FTIR)测定了3种缩叶藓属植物齿边缩叶藓、多枝缩叶藓和中华缩叶藓的红外谱图,运用离散小波变换对吸收较为相似的3种缩叶藓属植物的红外谱图进行特征提取。通过分析比较后选择第三,四分解层进行特征向量的提取,利用所得到的特征变量进行径向基神经网络(RBF-NN)训练,再将训练出来的网络对不同产地的3种缩叶藓属植物的红外谱图离散小波提取后的特征向量进行分类。通过对120个不同样本的验证,说明能够采用基于FTIR-离散小波进行数据压缩后进行特征变量的提取及径向基神经网络分类法对3种缩叶属植物齿边缩叶藓、多枝缩叶藓和中华缩叶藓进行分类。     

大鼠胰腺及癌组织红外光谱连续小波特征提取及径向基人工神经网络识别

采用水平衰减全反射(HATR)傅里叶变换红外光谱法(FTIR)测定了SD大鼠胰腺正常组织与非正常组织的谱图,提出了一种新的基于FTIR的连续小波特征提取与径向基人工神经网络分类方法以提高FTIR对早期SD大鼠胰腺癌的诊断准确率。利用连续小波多分辨率分析法提取FTIR特征量,对于提取的特征量采用径向基函数神经网络进行模式分类。对SD大鼠的胰腺正常组织、早期癌组织及进展期癌组织的FTIR,利用连续小波多分辨率分析法提取9个特征量,进行RBF神经网络分类判断。当目标误差为0.01,径向基函数的分布常数为5时,网络达到最优化,总的正确识别率为96.67%。并对影响分类结果的网络参数、目标误差和分布常数对分类样品的影响做了讨论。实验结果表明:此方法对早期胰腺癌具有较高的诊断率。     

尿液中常见毒品微量检测的表面增强拉曼光谱识别

通过将自适应平滑滤波器和结合小波变换的支持向量机(Support vector machine,SVM)分类器有机组合,建立了低信噪比拉曼光谱的模式识别方法。首先,通过自适应平滑滤波器进行光谱去噪,滤波窗口宽度根据信噪比估计值进行调整,从而在保证特征峰信号强度的同时达到更好的噪声滤波效果;其次,由小波变换实现光谱数据降维,通过小波分解层数优化可以获得训练集的最佳分类准确率;最后,由SVM进行分类,通过交叉验证(Cross validation,CV)实现SVM参数寻优,并根据交叉验证与分类器之间的准确率关系,得出分类器可用参数需满足的条件。基于表面增强拉曼光谱技术,本方法实现了人体尿液中甲基苯丙胺(Methamphetamine,MAMP)和亚甲基二氧基甲基苯丙胺(3,4-Methylenedio-xymethamphetamine,MDMA)的定性微量分析。实验使用中国科学院合肥智能机械研究所研发的金纳米棒拉曼光谱增强基底,由Delta Nu公司的Inspector型便携拉曼光谱仪采集光谱,激发光波长785 nm,曝光时间为5 s,整体检测准确率高于95.0%。     

小波变换-可见-近红外光谱技术用于鉴别品牌料酒的研究

模式识别技术广泛应用于食品种类、品牌和原产地的分类鉴别.本文测定了三个品牌114个料酒样品的可见-近红外光谱,利用小波变换技术对光谱信号进行了去噪和压缩处理,并采用Fisher权重法计算了16个小波细节系数的Fisher权重.以16个小波细节系数为特征变量采用向量相似度法对三种不同品牌料酒进行了相似度分析,主成分分析法...     

Comparison of diffuse-reflectance absorbance and attenuated total reflectance FT-IR for the discrimination of bacteria

A collection of bacteria belonging to the genus Bacillus were analysed by diffuse reflectance absorbance and attenuated total reflectance Fourier transform infrared (FT-IR) spectroscopy in the mid-infrared. The diffuse reflectance absorbance method is a rapid whole organism fingerprinting method, which generates a biochemical profile of the bacteria, where samples are presented to the FT-IR spectrometer dried on a metal carrier. The attenuated total reflectance FT-IR used in conjunction with a diamond attenuated total reflectance (ATR) accessory produces a biochemical profile of the surface chemistry of bacteria directly without the need for drying, and has not previously been used in the discrimination of bacteria. Principal component, discriminant function and hierarchical cluster analysis were performed on the data to discriminate the bacteria. The differentiation of the bacteria to species level was observed in both analyses however, it was concluded that the ATR FT-IR illustrated better sub-species differentiation of the microorganisms. This may imply that the total biochemical profiling infers discrimination to species level whereas strain specific markers are present on the cell surface chemistry.     

Fourier transform infrared (FT-IR) spectroscopy in bacteriology: towards a reference method for bacteria discrimination

Rapid and reliable discrimination among clinically relevant pathogenic organisms is a crucial task in microbiology. Microorganism resistance to antimicrobial agents increases prevalence of infections. The possibility of Fourier transform infrared (FT-IR) spectroscopy to assess the overall molecular composition of microbial cells in a non-destructive manner is reflected in the specific spectral fingerprints highly typical for different microorganisms. With the objective of using FT-IR spectroscopy for discrimination between diverse microbial species and strains on a routine basis, a wide range of chemometrics techniques need to be applied. Still a major issue in using FT-IR for successful bacteria characterization is the method for spectra pre-processing. We analyzed different spectra pre-processing methods and their impact on the reduction of spectral variability and on the increase of robustness of chemometrics models. Different types of the Enterococcus faecium bacterial strain were classified according to chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis (PFGE). Samples were collected from human patients. Collected FT-IR spectra were used to verify if the same classification was obtained. In order to further optimize bacteria classification we investigated whether a selected combination of the most discriminative spectral regions could improve results. Two different variable selection methods (genetic algorithms (GAs) and bootstrapping) were investigated and their relative merit for bacteria classification is reported by comparing with results obtained using the entire spectra. Discriminant partial least-squares (Di-PLS) models based on corrected spectra showed improved predictive ability up to 40% when compared to equivalent models using the entire spectral range. The uncertainty in estimating scores was reduced by about 50% when compared to models with all wavelengths. Spectral ranges with relevant chemical information for Enterococcus faecium bacteria discrimination were outlined.     

Identification of species’ blood by attenuated total reflection (ATR) Fourier transform infrared (FT-IR) spectroscopy

Blood is one of the most common and informative forms of biological evidence found at a crime scene. A very crucial step in forensic investigations is identifying a blood stain’s origin. The standard methods currently employed for analyzing blood are destructive to the sample and time-consuming. In this study, attenuated total reflection (ATR) Fourier transform infrared (FT-IR) spectroscopy is used as a confirmatory, nondestructive, and rapid method for distinction between human and animal (nonhuman) blood. Partial least squares-discriminant analysis (PLS-DA) models were built and demonstrated complete separation between human and animal donors, as well as distinction between three separate species: human, cat, and dog. Classification predictions of unknown blood donors were performed by the model, resulting in 100 % accuracy. This study demonstrates ATR FT-IR spectroscopy’s great potential for blood stain analysis and species discrimination, both in the lab and at a crime scene since portable ATR FT-IR instrumentation is commercially available.     

SFC/FT-IR: Supercritical fluid chromatography with Fourier transform infrared detection

Fourier transform infrared spectroscopy has been utilized in the discrimination of chromatographic eluents with success. The technique of capillary supercritical fluid chromatography (SFC) offers the possibility of highly efficient and quantitative separations which avoid problems associated with gas chromatography. The technique of SCF/FT-IR utilizing a flow-through cell is described. Approaches to solutions of several of the problems inherent in flow-through cell SFC/FT-IR are described.     

New Structures of Bismaleimides and Polyaspartimides with Ether Units

New bismaleimide monomers with various structures, synthesized by the reaction of maleic anhydride and new diamines, were used in the reaction with diamines to yield polyaspartimides. The structure of these monomers was confirmed by Fourier transform infrared (FT-IR) and proton nuclear magnetic resonance (¹H-NMR) spectroscopy, and polymer structure was evidenced by FT-IR spectroscopy. The resulting new compounds were characterized by differential scanning calorimetry (DSC) and thermogravimetry (TGA). The polymers exhibited film-forming ability. The quality of these films was studied by atomic force microscopy (AFM), and their mechanical properties (tensile strength, tensile modulus) were investigated.     

Fourier transform infrared microspectroscopy as a bacterial source tracking tool to discriminate fecal E. coli strains

The aim of this work was to use the FT-IR microspectroscopy technique, a union of a FT-IR spectrometer with a microscope, to discriminate fecal Escherichia coli strains from cows, chickens and humans and to compare the efficiency of this method with the genomic fingerprinting method, BOX-PCR. The obtained BOX-PCR profiles were able to correctly discriminate 93.75% of the chicken strains, 80% of the cow strains and 65% of the human strains. An efficient PLS-DA model was developed, using orthogonal signal correction and the second derivate of the FT-IR spectra. This model allowed the correct discrimination, according to the animal source, of all the E. coli strains analyzed. The bands in the FT-IR spectra that were responsible for the strains discrimination were in the region between 2816 and 3026 cm⁻¹ wavenumber, described as fatty acids. It was demonstrated that FT-IR microspectroscopy can be a suitable tool for fecal E. coli discrimination, because it is fast, easy to carry out and presents a flexible discrimination power.     

Applying Fourier-transform infrared spectroscopy and chemometrics to the characterization and identification of lactic acid bacteria

The infrared absorption spectra of 55 lactic acid bacteria belonging to the genera Lactobacillus, Weissella and Carnobacterium were obtained and mathematically analyzed. Sixteen reference strains and 39 food strains isolated from meat and meat products and belonging to the genera Lactobacillus (6 species), Weissella (3 species) and Carnobacterium (2 species) were processed under standardized conditions and their medium infrared spectra obtained using Fourier-transform infrared (FT-IR) spectroscopy. Reproducibility indexes and similarities between FT-IR spectra were calculated using modified correlation coefficients to detect the ranges with the best reproducibility and discrimination properties. Hierarchical cluster analysis and stepwise discriminant analysis (SDA) were subsequently carried out to detect classes and create library groups. Reference strains could be distinguished on the basis of their spectral data and their clustering was in agreement with differences in chemical composition of the cell wall. For the 39 food isolates, the capability of two identification systems was compared. Unknown strains were identified (a) using the linear functions obtained from SDA (canonical variables) of the variables that provide the best discrimination of spectra, and (b) by calculating a differentiation index when a range of the unknown's transformed IR spectrum was compared to all spectra included in a reference library. The system based on the differentiation index obtained a higher rate of identification, allowing for detection of outliers. FT-IR spectroscopy is shown to afford additional information to phenotypic and genotypic data which may help to establish a more robust taxonomic classification.     

Imaging of lipids in atherosclerotic lesion in aorta from ApoE/LDLR-/- mice by FT-IR spectroscopy and Hierarchical Cluster Analysis

Spectroscopy-based approaches can provide an insight into the biochemical composition of a tissue sample. In the present work Fourier transform infrared (FT-IR) spectroscopy was used to develop a reliable methodology to study the content of free fatty acids, triglycerides, cholesteryl esters as well as cholesterol in aorta from mice with atherosclerosis (ApoE/LDLR(-/-) mice). In particular, distribution and concentration of palmitic, oleic and linoleic acid derivatives were analyzed. Spectral analysis of pure compounds allowed for clear discrimination between free fatty acids and other similar moieties based on the carbonyl band position (1699-1710 cm(-1) range). In order to distinguish cholesteryl esters from triglycerides a ratio of carbonyl band to signal at 1010 cm(-1) was used. Imaging of lipids in atherosclerotic aortic lesions in ApoE/LDLR(-/-) mice was followed by Hierarchical Cluster Analysis (HCA). The aorta from C57Bl/6J control mice (fed with chow diet) was used for comparison. The measurements were completed with an FT-IR spectrometer equipped with a 128 × 128 FPA detector. In cross-section of aorta from ApoE/LDLR(-/-) mice a region of atherosclerotic plaque was clearly identified by HCA, which was later divided into 2 sub-regions, one characterized by the higher content of cholesterol, while the other by higher contents of cholesteryl esters. HCA of tissues deposited on normal microscopic glass, hence limited to the 2200-3800 cm(-1) spectral range, also identified a region of atherosclerotic plaque. Importantly, this region correlates with the area stained by standard histological staining for atherosclerotic plaque (Oil Red O). In conclusion, the use of FT-IR and HCA may provide a novel tool for qualitative and quantitative analysis of contents and distribution of lipids in atherosclerotic plaque.     

Pollen discrimination and classification by Fourier transform infrared (FT-IR) microspectroscopy and machine learning

The discrimination and classification of allergy-relevant pollen was studied for the first time by mid-infrared Fourier transform infrared (FT-IR) microspectroscopy together with unsupervised and supervised multivariate statistical methods. Pollen samples of 11 different taxa were collected, whose outdoor air concentration during the flowering time is typically measured by aerobiological monitoring networks. Unsupervised hierarchical cluster analysis provided valuable information about the reproducibility of FT-IR spectra of the same taxon acquired either from one pollen grain in a 25 × 25 μm² area or from a group of grains inside a 100 × 100 μm² area. As regards the supervised learning method, best results were achieved using a K nearest neighbors classifier and the leave-one-out cross-validation procedure on the dataset composed of single pollen grain spectra (overall accuracy 84%). FT-IR microspectroscopy is therefore a reliable method for discrimination and classification of allergenic pollen. The limits of its practical application to the monitoring performed in the aerobiological stations were also discussed. Figure Traditional and innovative methods for the identification of airborne pollen grains