近红外光谱-径向基神经网络在异烟肼片无损定量分析中的应用

应用异烟肼片粉末的近红外漫反射光谱数据分别结合偏最小二乘法(PLS)和径向基神经网络(RBFNN)建立定量分析模型,并用所建模型对预测集样品进行了预测,结果表明:应用RBFNN所建立的定量分析模型优于PLS模型,相关系数(r)值由0.99593提高到0.99734,交互验证均方根误差(RMSECV)值由0.00523下降到0.00423,预测均方根误差(RMSEP)值由0.00614下降到0.00501。     

近红外漫反射光谱法测定黄连浸膏粉中生物碱含量

采用近红外漫反射光谱法对黄连浸膏粉中小檗碱、巴马亭、药根碱和总生物碱含量进行快速无损检测。以HPLC分析值作参比,采用偏最小二乘回归算法建立二阶导数光谱信息与各组分含量间的定量校正模型,并对未知样品中各组分含量进行预测。小檗碱、巴马亭、药根碱和总生物碱的预测均方差(RMSEP)分别为0．184、0．109、0．054和0．325;加样回收率分别为97．67％～99．59％、96．63％～100．70％、95．15％～101．15％和97．41％～99．89％,重现性实验相对标准偏差(RSD)分别为0．3％、0．6％、1．8％和0．3％。该方法结果准确可靠,重现性、稳定性均良好,适用于工业现场的原位和在线检测。     

Quantitative Analysis of Berberine in Processed Coptis by Near-Infrared Diffuse Reflectance Spectroscopy

The near-infrared（NIR） diffuse reflectance spectroscopy was used to study the content of Berberine in the processed Coptis. The allocated proportions of Coptis to ginger, yellow liquor or Evodia rutaecarpa changed according to the results of orthogonal design as well as the temperature. For as withdrawing the full and effective information from the spectral data as possible, the spectral data was preprocessed through first derivative and multiplicative scatter correetion（MSC） according to the optimization results of different preprocessing methods. Firstly, the model was established by partial least squares（PLS）; the coefficient of determination（R2） of the prediction was 0.839, the root mean squared error of prediction（RMSEP） was 0.1422, and the mean relative error（RME） was 0.0276. Secondly, for reducing the dimension and removing noise, the spectral variables were highly effectively compressed via the wavelet transformation（WT） technology and the Haar wavelet was selected to decompose the spectral signals. After the wavelet coefficients from WT were input into the artificial neural network（ANN） instead of the spectra signal, the quantitative analysis model of Berberine in processed Coptis was established. The R^2 of the model was 0.9153, the RMSEP was 0.0444, and the RME was 0.0091. The values of appraisal index, namely R^2, RMSECV, and RME, indicate that the generalization ability and prediction precision of ANN are superior to those of PLS. The overall results show that NIR spectroscopy combined with ANN can be efficiently utilized for the rapid and accurate analysis of routine chemical compositions in Coptis. Accordingly, the result can provide technical support for the further analysis of Berberine and other components in processed Coptis. Simultaneously, the research can also offer the foundation of quantitative analysis of other NIR application.     

Enhanced Characterization of Naproxen Formulation by Near Infrared Spectroscopy

Near infrared spectroscopy in combination with appropriate chemometric methods is an effective technique for quantitative analysis of parameters of interest for the pharmaceutical industry. In this study, the artificial neural network (ANN) was applied to monitor critical parameters (compression force, tablet hardness, mean particle size, and active pharmaceutical ingredient concentration of tablets) in the process of naproxen pharmaceutical preparation. The performance of ANN was compared to linear methods (partial least squares regression (PLS) and synergy interval partial squares (siPLS)). The ANN models for compression force, tablet hardness, mean particle size, and active pharmaceutical ingredient concentration of tablets yielded the low root mean square error of prediction (RMSEP) values of 0.936 KN, 0.302 kg, 4.49 mg, and 2.14 µm, respectively. The predictive ability of the PLS model was improved by siPLS with selection of spectral regions and the best performance among all calibration methods was showed by the nonlinear method (ANN). Effective models were built by using these approaches using near infrared spectroscopy.     

Determination of Bovine Hemoglobin by Near-Infrared Diffuse Reflectance Spectroscopy with Novel Adsorption Preconcentration

A method for the quantitative determination of bovine hemoglobin in dilute solution was developed using adsorption preconcentration and near-infrared diffuse reflectance spectroscopy. An adsorbent, designated as multicarbonyl polymer-grafted silica particles, was prepared for the preconcentration of bovine hemoglobin in dilute solution. Under neutral conditions, the adsorption efficiency exceeded 98% within 20?min. After the preconcentration of bovine hemoglobin on the adsorbent, the near-infrared spectrum was measured in diffuse reflectance mode and a partial least squares model was constructed for quantitative prediction. Samples were analyzed in the presence of amino acid, albumin bovine V, D-glucose, and metal ions as potential interferences. The results show that bovine hemoglobin was selectively determined. The correlation coefficient between the predicted concentrations and the reference values was 0.9911, and the recoveries were from 86.4 to 111.2% for validation samples with concentrations between 2.1 and 30.0?mg?L^?1. Therefore, the determination of bovine hemoglobin was achieved by near-infrared diffuse reflectance spectroscopy combined with preconcentration and chemometric modeling.     

Determination of Selenium by Single-Drop Microextraction and Diffuse Reflectance Infrared Spectroscopy

A simple and convenient assay based on single-drop microextraction with infrared spectroscopy is reported for the determination of selenium. The extraction conditions were carefully optimized and selenium was preconcentrated through single-drop microextraction in 1,2-dichloroethane containing N-hydroxy-N-phenyl-N′-(o-tolyl) benzimidamide. The method is selective and almost all common ions including molybdenum(VI), chromium(VI), and tungsten(VI) did not interfere with the isolation protocol. The selenite band at 875?±?2?cm^?1, which is assigned to the asymmetric vibrational stretch (υ₃), was used for the quantification of selenium. Low limits of detection and quantification of 2.0 and 6.6?µg?L^?1 demonstrate the sensitivity of the method. Good precision was evaluated by the standard deviation (2.0?µg?L^?1) and relative standard deviation (0.5%) for 8?µg?L^?1 was achieved for 10 measurements. The method was used to analyze human blood, urine, and water for selenium.     

用人工神经网络-近红外光谱法测定冬虫夏草中的甘露醇

提出了用近红外漫反射光谱技术快速分析发酵冬虫夏草菌粉中甘露醇含量的新方法。采用比色法测定样品中的甘露醇,其含量范围为8．082％-14．548％。在7501．7-6097．8cm^-1与5453．7-4246．5cm^-1波段,分别采用PCR、PLSR和BP神经网络方法建立了样品近红外光谱的一阶微分光谱与其甘露醇含量之间的相关模型。BP神经网络模型的内部交叉验证误差均方根为0．475,预测误差均方根为0．608,均优于PCR和PLSR的处理结果。这表明,BP神经网络法对非线性检测对象具有较好的建模效果,可用于中药近红外光谱分析的非线性校正。     

中红外、近红外和拉曼光谱法测定商品农药制剂中氰戊菊酯和马拉硫磷的含量

利用近红外、中红外和拉曼光谱法定量分析了商品农药制剂中有效成分氰戊菊酯和马拉硫磷的含量.采用偏最小二乘法(Partial least squares,PLS)建立氰戊菊酯和马拉硫磷的定量模型并进行了优化,用独立检验集对模型适应性进行评价.近红外和中红外法测定氰戊菊酯、马拉硫磷定量模型的相关系数分别是0.9981,0.9994和0.9946,0.9998,外部验证集标准差分别是0.082,0081和0.092,0.075,两种方法的定量效果接近;拉曼法氰戊菊酯和马拉硫磷定量模型的相关系数分别为0.9872和0.9993,外部验证集标准差分别为0.254和0.317,预测精度不及近红外和中红外法高.MIR-ATR,NIR和Raman 3种方法均能满足现场检测农药质量的需要.     

径向基函数网络在近红外人体无创伤血糖浓度检测基础研究中的应用

在近红外无创伤血糖浓度检测的基础研究中,对于多组分的混合物的分析,常因光谱与样品浓度之间呈现非线性响应,使得基于线性模型的校正方法失效。本文讨论了非线性校正方法径向基函数神经网络( RBFN )的有效性,并与线性校正方法中的主成分分析和偏最小二乘法作了对比研究。验证实验所用样品为①葡萄糖水溶液②包含牛血红蛋白和白蛋白的葡萄糖水溶液,结果表明：在①实验中PLS模型和RBFN预测标准偏差分别为8．2、8．9;在②实验中分别为15．6、8．8。可见在样品组分增多时,RBFN算法较线性PLS方法建立的模型预测能力强。     

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。     

径向基函数神经网络荧光光度法同时测定痕量铝、镓、铟、铊

以径向基网络(RBF)对荧光光谱严重重叠的Al3 、Ga3 I、n3 、Tl3 四组分混合体系同时进行测定。通过正交设计安排样本,在激发波长390 nm下,测定446~615nm的发射光谱。以34个特征波长处的荧光强度值作为网络特征参数,经网络训练和计算得出Al3 、Ga3 、In3 、Tl3 四者的平均回收率分别为99.07%、103.49%、98.72%、95.04%,在时间和精度上都比LMBP网络优越。     

烟草灰分、总挥发酸和总挥发碱的近红外光谱分析

应用偏最小二乘法(PLS)结合近红外光谱(NIR)对烟草灰分(ash)、总挥发酸(TVA)和总挥发碱(TVB)建立校正模型。烟草灰分、总挥发酸和总挥发碱模型相关系数分别为0.97312、0.96220和0.98050;均方预测残差(RMSECV)分别为0.41227、0.00688和0.09790;预测范围分别为1.74~31.31、0.0570~0.2336和0.042~1.136;通过对模型进行t-检验,在显著性水平大于0.05的条件下,其预测结果与行业标准方法的测定结果对比,结果令人满意。     

应用光谱技术无损检测油菜叶片中乙酰乳酸合成酶

应用可见/近红外光谱技术实现了油菜叶片中乙酰乳酸合成酶(ALS)的快速无损检测.对99个油菜样本进行光谱扫描,经过平滑、变量标准化、一阶求导等预处理后,应用偏最小二乘法(PLS)建立了ALS的预测模型.同时提取有效特征变量,作为反向传输人工神经网络(BPNN)和最小二乘-支持向量机(LS-SVM)的输入值,并建立相应的模型.用66个样本建模,33个样本验证.结果表明,LS-SVM模型能够获得最优的预测结果,预测集样本的相关系数(r)、预测标准差(RMSEP)和偏差(Bias)分别为0.998、 0.715和0.079,获得了满意的预测精度.结果表明,应用可见/近红外光谱技术结合LS-SVM检测油菜中乙酰乳酸合成酶是可行的,并能获得满意的预测精度,为进一步应用光谱技术进行油菜生长状况的大田监测奠定了基础.     

近红外光谱法快速测定生物柴油调和比及理化指标

通过比较偏最小二乘法(PLS)处理调和生物柴油近红外光谱图与标准方法测定调和生物柴油所获得的基础数据,确立了调和生物柴油的调和比、密度、运动黏度、热值、闭口闪点及冷凝点之间的相互关系。结果表明:经优化后,在OPUS光谱分析软件推荐维数(Rank)下,各指标模型的预测值与标准测定值之间线性相关关系均显著。在用于测定未知调和生物柴油样品的上述指标方面具有测定快速简便、误差小、成本低等优点,并用马氏距离限制异常项,每份生物柴油各指标的马氏距离都处于允许范围内。对于新类型生物柴油,可向模型添加10个以上调配样本,扩充模型后即可用于测定这类型调和生物柴油相关理化指标,可成为测定调和生物柴油相关理化指标新方法。在此基础上,可进一步开发出生物柴油近红外光学理化指标测定仪,实现低成本与快速测定。     

膜富集-可见漫反射光谱法测定池塘水和维生素B_(12)药片中痕量钴

1-(2-吡啶偶氮)-2-萘酚与钴离子生成暗绿色的络合物,反应后将其过滤到膜上,测得膜表面漫反射光谱,据此提出了膜富集-可见漫反射光谱法测定池塘水和维生素B12药片中痕量钴的方法。钴的质量浓度在0.2~3.0μg·L-1和3.0~10.0μg·L-1范围内与其吸光度呈线性关系,检出限(3s/k)为0.093μg·L-1。方法用于池塘水和维生素B12药片的分析,加标回收率在96.7%~105%之间,测定值的相对标准偏差在1.4%~2.9%之间。     

偏最小二乘-反向传播-近红外光谱法同时测定饲料中4种氨基酸

偏最小二乘与人工神经网络联用对70个饲料样品建立起天门冬氨酸(Asp)、谷氨酸(Glu)、丝氨酸(Ser)和组氨酸(His)4种氨基酸含量的预测校正模型,以样品平行扫描光谱验证校正模型预测的准确性和重现性。用偏最小二乘法将原始数据压缩为主成分,采用单隐层的反向传播网络建模。取前3个主成分的12个数据输入网络,以Kolmogorov定理为依据,经过实验确定中间层的神经元个数为25,初始训练迭代次数为1000。偏最小二乘-反向传播网络模型对样品4个组分含量的预测决定系数(R2)分别为:0.981、0.997、0.979、0.946;样品平行扫描光谱预测值的标准偏差分别为:0.020、0.029、0.017、0.023。本研究为近红外快速检测在组分含量较低的样品实现多组分同时测定提供了思路。     

模糊聚类-RBF神经网络法同时测定铁、镍、钒、铜、钴

将模糊聚类分析与RBF神经网络法相结合,对模拟原油中吸收光谱严重重叠的重金属多组分体系进行解析,较好地解决了光度分析计算中校准模型的优化问题,提高了分析结果的准确度。用此方法研究铁、镍、钒、铜、钴-乳化剂OP-5-Br-PADAP螯合显色体系,计算表明模拟样品各组分的回收率为94%~105.5%,油样测定结果的相对标准偏差在1.85%~4.3%之间。     

紫外可见吸收光谱结合主成分-反向传播人工神经网络鉴别真假蜂蜜

研究紫外-可见吸收光谱技术结合化学计量学方法鉴别真假蜂蜜.根据蜂蜜中果糖和葡萄糖的典型质量比1.2:1.0,配制与真蜂蜜相近的掺假溶液,并以5%～20%的比例掺入真蜂蜜中.获取纯正蜂蜜和掺假蜂蜜的紫外-可见吸收光谱,选择最佳敏感波段250～400 mn的吸光度值进行主成分分析(PCA),优选主成分作为反向传播人工神经网...