Yulin Ren

Diagnosis of lung cancer based on metal contents in serum and hair using multivariate statistical methods.

The classification of normal and cancer groups with four multivariate methods according to metal contents in serum and hair samples has been discussed in the present paper. Results show that the four multivariate methods, stepwise discrimination analysis, principal components analysis, hierarchical cluster analysis, and stepwise cluster analysis can distinguish the two groups correctly. The independent samples of both normal and cancer groups were tested and can be distinguished correctly by the four methods. Therefore, these methods can be used as an aid for diagnosis of lung cancer according to the metal contents in serum and hair samples.

Nondestructive Quantitative Analysis of Analgini Powder Pharmaceutical by Near-Infrared Spectroscopy and Artifical Neural Network Technique

ABSTRACT This paper demonstrates the usefulness of near-infrared (NIR) spectra and artificial neural network (ANN) in nondestructive quantitative analysis of pharmaceuticals. Real data sets from near-infrared reflectance spectra of analgini powder pharmaceutical were used to build up an artificial neural network to predict unknown samples. The parameters affecting the network were discussed. A new network evaluation criterion, the degree of approximation, was employed. The overfitting was discussed. Owing to the good nonlinear multivariate calibration nature of ANN, the predicted result was reliable and precise. The relative error of unknown samples was less than 2.5%

Nondestructive Quantitative Analysis of Cimetidine Tablets Using Artificial Neural Networks in Near‐Infrared Spectroscopy

Abstract This project was designed to explore the application of artificial neural networks (ANNs) and near‐infrared (NIR) spectroscopy for nondestructive quantitative analysis of cimetidine tablets. The models of conventional spectra (SNV pretreated), first‐derivative spectra, and second‐derivative spectra, have been established, respectively. In order to be compared with the tablets, the powders were also determined. Both tablets and powders were found to provide similar results in the quantification of the active compound (cimetidine). Of all the models, the second‐derivative models resulted in the lowest relative standard error (<0.1%). The parameters affecting the network were discussed, and unknown specimens were predicted. The degree of approximation, a new evaluation criterion of the network, was employed, which proved the accuracy of the predicted results.

Quantitative analysis of diclofenac sodium powder via near-infrared spectroscopy combined with artificial neural network.

A method for quantitative analysis of diclofenac sodium powder on the basis of near-infrared (NIR) spectroscopy is investigated by using of orthogonal projection to latent structures (O-PLS) combined with artificial neural network (ANN). 148 batches of different concentrations diclofenac sodium samples were divided into three groups: 80 training samples, 46 validation samples and 22 test samples. The average concentration of diclofenac sodium was 27.80%, and the concentration range of all the samples was 15.01-40.55%. O-PLS method was applied to remove systematic orthogonal variation from original NIR spectra of diclofenac sodium samples, and the filtered signal was used to establish ANN model. In this model, the concentration of diclofenac sodium was determined. The degree of approximation was employed as selective criterion of the optimum network parameters. In order to compare with O-PLS-ANN model, principal component artificial neural network (PC-ANN) model and calibration models that use different preprocessing methods (first derivative, standard normal variate (SNV) and multiplicative scatter correction (MSC)) of the original spectra were also designed. In addition, partial least squares regression (PLS) models were also established to compare with ANN models. Experimental results show that O-PLS-ANN model is the best.

Application of Artificial Neural Network Multivariate Calibration to Near-Infrared Spectrophotometry Determination of Powdered Pharmaceutical Metronidazole

Abstract The application of artificial neural networks for pharmaceutical quantitative analysis is described. Real data sets from near-infrared reflectance spectra of Metronidazole powdered pharmaceuticals were used to build an artificial neural network to predict unknown samples. A new network evaluation criterion, termed the degree of approximation, was employed. The overfitting was discussed. Owing to the beneficial nonlinear multivariate calibration nature of ANN. the predicted results were reliable and precise.

Quantitative analysis of cefalexin based on artificial neural networks combined with modified genetic algorithm using short near-infrared spectroscopy

In this paper, a novel chemometric method was developed for rapid, accurate, and quantitative analysis of cefalexin in samples. The experiments were carried out by using the short near-infrared spectroscopy coupled with artificial neural networks. In order to enhancing the predictive ability of artificial neural networks model, a modified genetic algorithm was used to select fixed number of wavelength.

Application of artificial neural networks to the nondestructive determination of ciprofloxacin hydrochloride in powder by short-wavelength NIR spectroscopy

The present study is aimed at providing a new short-wavelength near-infrared (NIR) spectroscopic method for the nondestructive quantitative analysis of ciprofloxacin hydrochloride in powder via artificial neural networks (ANNs). For this purpose, the NIR spectra of 90 experimental powder samples in the range 700–1100 mm were analyzed. Four different pretreatment methods—first-derivative, second-derivative, standard normal variate (SNV), and multiplicative scatter correction (MSC)—were applied to three sets of the NIR spectra of the powder samples. Among all of the ANN models, the first-derivative model is found to be the best. The results presented here demonstrate that the short-wavelength NIR region is promising for the fast and reliable determination of the major components in pharmaceuticals. The degree of approximation as an evaluation criterion prevents the overfitting phenomenon occurring in ANNs.

Quality Control of Pharmaceutical Aspirin Powder by Near-Infrared Reflectance Spectra and Multivariate Analysis

Abstract Quality control of pharmaceutical aspirin powder was studied using first order differential near-infrared diffuse reflectance spectra and four standard multivariate methods, hierarchical clustering analysis, stepwise clustering analysis, principal components analysis, and stepwise discrimination. The qualified, inferior, and fake pharmaceutical aspirin powders of independent samples can be distinguished by the multivariate analysis methods based on the reflectance spectra. The proposed methods are reliable, fast and nondestructive.

Orthogonal projection to latent structures combined with artificial neural networks in non-destructive analysis of Ampicillin powder.

A new method orthogonal projection to latent structures (O-PLS) combined with artificial neural networks is investigated for non-destructive determination of Ampicillin powder via near-infrared (NIR) spectroscopy. The modern NIR spectroscopy analysis technique is efficient, simple and non-destructive, which has been used in chemical analysis in diverse fields. Be a preprocessing method, O-PLS provides a way to remove systematic variation from an input data set X not correlated to the response set Y, and does not disturb the correlation between X and Y. In this paper, O-PLS pretreated spectral data was applied to establish the ANN model of Ampicillin powder, in this model, the concentration of Ampicillin as the active component was determined. The degree of approximation was employed as the selective criterion of the optimum network parameters. In order to compare the OPLS-ANN model, the calibration models that using first-derivative and second-derivative preprocessing spectra were also designed. Experimental results showed that the OPLS-ANN model was the best.