Gongju(Chrysanthemum morifolium) is one of the five major medicinal Chrysanthemum varieties included in the Chinese Pharmacopoeia. In recent years, its cultivation areas have changed significantly, resulting in mixed quality of the medicinal herbs. In this study, Gongju cultivated in Anhui, Yunnan, Chongqing, and other places were selected as research objects. Hyperspectral data were collected in the visible-near-infrared(VNIR) and short-wave infrared(SWIR) bands using different modes, such as corolla facing up(A) and flower base facing up(B). After pre-processing the hyperspectral data using five methods, including multiplicative scatter correction(MSC), Savitzky-Golay smoothing(SG), first derivative(D1), second derivative(D2), and standard normal variate(SNV), partial least squares discriminant analysis(PLSDA), random forest(RF), and support vector machine(SVM) were used to establish origin identification models of Gongju at the two geographical scales of the province and the city-county in Anhui province. The accuracy of the prediction results was used as an evaluation index to select the optimal models, and the classification performance of the models was evaluated by confusion matrix. The results showed that the flower base facing up(B) collection model combined with second derivative pretreatment and RF method was the best model for both geographical scale identification models. The modeling effect of the full-band(VNIR + SWIR) was slightly better than that of the single band, with the accuracy of the prediction set in the province and city-county regions reaching 99.69% and 99.40%, respectively. The competitive adaptive reweighted sampling algorithm(CARS), successive projections algorithm(SPA), and variable iterative space shrinkage approach(VISSA) were further used to screen the feature wavelength modeling. The number of feature wavelengths screened by CARS was fewer, and the prediction set accuracy of the two geographical scales models after optimization could reach 99.56% and 98.65%, which was basically comparable to the full-band model. However, the removal of redundant variables could greatly reduce the complexity of the model. The hyperspectral technology combined with the chemometrics model established in this study can achieve the origin identification of Gongju at different geographical scales, providing a theoretical basis and technical reference for the construction of a rapid detection system for Gongju origin and the development of exclusive miniaturized instrumentation and equipment systems.
Chrysanthemum/growth & development* ; China ; Support Vector Machine ; Geography ; Discriminant Analysis ; Spectroscopy, Near-Infrared/methods* ; Spectrum Analysis/methods* ; Drugs, Chinese Herbal/analysis* ; Least-Squares Analysis

This study aims to establish a rapid and non-destructive method for recognizing the origins and cultivation patterns of Astragali Radix. A hyperspectral imaging system(spectral ranges: 400-1 000 nm, 900-1 700 nm; detection time: 15 s) was used to examine the samples of Astragali Radix with different origins and cultivation patterns. The collected hyperspectral datasets were highly correlated and numerous, which required the establishment of stable and reliable dimension reduction and classification models. Firstly, the original spectra were preprocessed by normalization, Gaussian smoothing, and masking. Then, principal component analysis(PCA), partial least squares-discriminant analysis(PLS-DA), and competitive adaptive reweighted sampling(CARS) were performed to reduce the dimension of the hyperspectral data. Finally, support vector machine(SVM), feedforward neural network(FFNN), and convolutional neural network(CNN) were used for data training of the spectral images and spectral curves with dimension reduction. The results showed that applying CARS as a variable selection method before PLS-DA on the hyperspectral data of Astragali Radix achieved the accuracy, precision, and recall of 100% on the CNN test dataset. The F_1-score and area under the curve of ROC(AUC) reached 1. This method is convenient, quick, sample-saving, and non-destructive, providing technical support for rapid identification of the origins and cultivation patterns of Astragali Radix.
Drugs, Chinese Herbal/chemistry* ; Neural Networks, Computer ; Algorithms ; Support Vector Machine ; Principal Component Analysis ; Discriminant Analysis ; Hyperspectral Imaging/methods* ; Least-Squares Analysis ; Astragalus Plant/growth & development* ; Astragalus propinquus/growth & development*

This Fructus,study including and aimed to construct a rapid and nondestructive detection flavonoid,model betaine,for and of the content vitamin of(Vit four four quality C).index components Lycium barbarum polysaccharide,of inL ycii rawma total and C Hyperspectral data quantitative of terials modelswere powder developed Lycii using Fructus partial were squares effects collected,regression raw based LSR),on the support content vector the above components,the forest least(P regression compared,(SVR),the and effects random three regression(RFR)were algorithms.also The Four spectral predictive commonly data of the materialsand powder were were applied and of spectral quantitative for models reduction.compared.used were pre-processing screened methods feature to successive pre-process projection the raw algorithm data(SPA),noise competitive Thepre-processed for bands using adaptive reweigh ted sampling howed(CARS),the and maximal effects relevance based and raw minimal materials redundancy and(MRMR)were algorithms Following to optimize multiplicative the models.scatter The correction Based resultss(MS that prediction SPA on feature the powder prediction similar.PLSR C)denoising sproposed and integrated for model,screening the the coefficient bands,determination the effect(R_C~2)of(MSC-SPA-PLSR)coefficient was optimal.of on(R_P~2)thi of of calibration flavonoid,and and of all determination greater prediction0.83,L.barbarum inconte nt prediction of polysaccharide,total mean betaine,of Vit C were than smallest In the compared study,root with mean other prediction content squareserror models of the calibration(RMSEC)residual and deviation root squares was error2.46,prediction2.58,(RMSEP)and were the,and prediction(RPD)2.50,developed3.58,achieve respectively.rapid this the the quality mod el(MSC-SPA-PLSR)fourcomponents based Fructus,on hyperspectral which technology was approach to rapid and effective detection detection of the of Lycii in Lycii provided a new to the and nondestructive of of Fructus.
Spectroscopy, Near-Infrared/methods* ; Betaine ; Powders ; Least-Squares Analysis ; Algorithms ; Flavonoids

To realize the non-destructive and rapid origin discrimination of Poria cocos in batches, this study established the P. cocos origin recognition model based on hyperspectral imaging combined with machine learning. P. cocos samples from Anhui, Fujian, Guangxi, Hubei, Hunan, Henan and Yunnan were used as the research objects. Hyperspectral data were collected in the visible and near infrared band(V-band, 410-990 nm) and shortwave infrared band(S-band, 950-2 500 nm). The original spectral data were divided into S-band, V-band and full-band. With the original data(RD) of different bands, multiplicative scatter correction(MSC), standard normal variation(SNV), S-G smoothing(SGS), first derivative(FD), second derivative(SD) and other pretreatments were carried out. Then the data were classified according to three different types of producing areas: province, county and batch. The origin identification model was established by partial least squares discriminant analysis(PLS-DA) and linear support vector machine(LinearSVC). Finally, confusion matrix was employed to evaluate the optimal model, with F1 score as the evaluation standard. The results revealed that the origin identification model established by FD combined with LinearSVC had the highest prediction accuracy in full-band range classified by province, V-band range by county and full-band range by batch, which were 99.28%, 98.55% and 97.45%, respectively, and the overall F1 scores of these three models were 99.16%, 98.59% and 97.58%, respectively, indicating excellent performance of these models. Therefore, hyperspectral imaging combined with LinearSVC can realize the non-destructive, accurate and rapid identification of P. cocos from different producing areas in batches, which is conducive to the directional research and production of P. cocos.
Hyperspectral Imaging ; Wolfiporia ; China ; Least-Squares Analysis ; Support Vector Machine

In this study, visual-near infrared(VNIR), short-wave infrared(SWIR), and VNIR + SWIR fusion hyperspectral data of Polygonatum cyrtonema from different geographical origins were collected and preprocessed by first derivative(FD), second derivative(SD), Savitzky-Golay smoothing(S-G), standard normalized variate(SNV), multiplicative scatter correction(MSC), FD+S-G, and SD+S-G. Three algorithms, namely random forest(RF), linear support vector classification(LinearSVC), and partial least squares discriminant analysis(PLS-DA), were used to establish the identification models of P. cyrtonema origin from three spatial scales, i.e., province, county, and township, respectively. Successive projection algorithm(SPA) and competitive adaptive reweighted sampling(CARS) were used to screen the characteristic bands, and the P. cyrtonema origin identification models were established according to the selected characteristic bands. The results showed that(1)after FD preprocessing of VNIR+SWIR fusion hyperspectral data, the accuracy of recognition models established using LinearSVC was the highest, reaching 99.97% and 99.82% in the province origin identification model, 100.00% and 99.46% in the county origin identification model, and 99.62% and 98.39% in the township origin identification model. The accuracy of province, county, and township origin identification models reached more than 98.00%.(2)Among the 26 characteristic bands selected by CARS, after FD pretreatment, the accuracy of origin identification models of different spatial scales was the highest using LinearSVC, reaching 98.59% and 97.05% in the province origin identification model, 97.79% and 94.75% in the county origin identification model, and 90.13% and 87.95% in the township origin identification model. The accuracy of identification models of different spatial scales established by 26 characteristic bands reached more than 87.00%. The results show that hyperspectral imaging technology can realize accurate identification of P. cyrtonema origin from different spatial scales.
Spectroscopy, Near-Infrared ; Polygonatum ; Algorithms ; Random Forest ; Least-Squares Analysis

This study was aimed at identifying the bioactive components of the crude and stir-baked hawthorn for invigorating spleen and promoting digestion, respectively, to clarify the processing mechanism of hawthorn by applying the partial least squares(PLS) algorithm to build the spectrum-effect relationship model. Firstly, different polar fractions of crude and stir-baked hawthorn aqueous extracts and combinations of different fractions were prepared, respectively. Then, the contents of 24 chemical components were determined by ultra-high performance liquid chromatography-mass spectrometry. The effects of different polar fractions of crude hawthorn and stir-baked hawthorn aqueous extracts and combinations of different fractions were evaluated by measuring the gastric emptying rate and small intestinal propulsion rate. Finally, the PLS algorithm was used to establish the spectrum-effect relationship model. The results showed that there were significant differences in the contents of 24 chemical components for different polar fractions of crude and stir-baked hawthorn aqueous extracts and combinations of different fractions, and the gastric emptying rate and small intestinal propulsion rate of model rats were improved by administration of different polar fractions of crude and stir-baked hawthorn aqueous extracts and combinations of different fractions. The bioactive components of crude hawthorn identified by PLS models were vitexin-4″-O-glucoside, vitexin-2″-O-rhamnoside, neochlorogenic acid, rutin, gallic acid, vanillic acid, citric acid, malic acid, quinic acid and fumaric acid, while neochlorogenic acid, cryptochlorogenic acid, rutin, gallic acid, vanillic acid, citric acid, quinic acid and fumaric acid were the bioactive components of stir-baked hawthorn. This study provided data support and scientific basis for identifying the bioactive components of crude and stir-baked hawthorn, and clarifying the processing mechanism of hawthorn.
Animals ; Rats ; Spleen ; Crataegus ; Quinic Acid ; Least-Squares Analysis ; Vanillic Acid ; Algorithms ; Digestion

This study established a method for rapid quantification of terpene lactone, bilobalide, ginkgolide C, ginkgolide A and ginkgolide B in the chromatographic process of Ginkgo Folium based on near infrared spectroscopy(NIRS). The effects of competitive adaptive reweighting sampling(CARS), random frog(RF), and synergy interval partial least squares(siPLS) on the performance of partial least squares regression(PLSR) model were compared to the reference values measured by HPLC. Among them, the correlation coefficients of prediction(Rp) of validation sets of terpene lactone, bilobalide, and ginkgolide C were all higher than 0.98, and the relative standard errors of prediction(RSEPs) were 5.87%, 6.90% and 6.63%, respectively. Aiming at ginkgolide A and ginkgolide B with relatively low content, the genetic algorithm joint extreme learning machine(GA-ELM) was used to establish the optimized quantitative analysis model. Compared with CARS-PLSR model, the CARS-GA-ELM models of ginkgolide A and ginkgolide B exhibited a reduction in RSEP from 15.65% to 8.52% and from 21.28% to 10.84%, respectively, which met the needs of quantitative ana-lysis. It has been proved that NIRS can be used for the rapid detection of various lactone components in the chromatographic process of Ginkgo Folium.
Chromatography, High Pressure Liquid ; Ginkgo biloba ; Lactones/analysis* ; Least-Squares Analysis ; Spectroscopy, Near-Infrared/methods*

In order to realize the rapid and non-destructive detection of SO_2 content in sulphur-fumigated Achyranthis Bidentatae Radix, this paper first prepared the sulphur-fumigated Achyranthis Bidentatae Radix samples with the usage amount of sulphur being 0, 2.5%, and 5% of the mass of Achyranthis Bidentatae Radix pieces. The SO_2 content in different batches of sulphur-fumigated Achyranthis Bidentatae Radix was determined using the method in Chinese Pharmacopoeia, followed by the acquisition of their hyperspectral data within both visible-near infrared(435-1 042 nm) and short-wave infrared(898-1 751 nm) regions by hyperspectral imaging. Meanwhile, the first derivative, AUTO, multiplicative scatter correction, Savitzky-Golay(SG) smoothing, and standard normal variable transformation algorithms were used to pre-process the original hyperspectral data, which were then subjected to characteristic band extraction based on competitive adaptive reweighted sampling(CARS) and the partial least square regression analysis for building a quantitative model of SO_2 content in sulphur-fumigated Achyranthis Bidentatae Radix. It was found that the accuracy of the quantitative model built depending on the visible-near infrared spectra was high, with the determination coefficient of prediction set(R■) reaching 0.900 1. The established quantitative model has enabled the rapid and non-destructive detection of SO_2 content in sulphur-fumigated Achyranthis Bidentatae Radix, which can serve as an effective supplement to the method described in Chinese Pharmacopeia.
Hyperspectral Imaging ; Least-Squares Analysis ; Plant Roots ; Sulfur

In this study, 280 batches of Zingiberis Rhizoma samples from nine producing areas were analyzed to obtain infrared spectral information based on near-infrared spectroscopy(NIRS). Pluralistic chemometrics such as principal component analysis(PCA), partial least squares-discriminant analysis(PLS-DA), orthogonal partial least squares-discriminant analysis(OPLS-DA), K-nearest neighbors(KNN), support vector machine(SVM), random forest(RF), artificial neural network(ANN), and gradient boosting(GB) were applied for tracing of origins. The results showed that the discriminative accuracy of the spectral preprocessing by standard normal variate transformation coupled with the first derivative was 93.9%, which could be used for the construction of the discrimination model. PCA and PLS-DA score plots showed that samples from Shandong, Sichuan, Yunnan, and Guizhou could be effectively distinguished, but the remaining samples were partially overlapped. As revealed by the analysis results by machine learning algorithms, the AUC values of KNN, SVM, RF, ANN, and GB algorithms were 0.96, 0.99, 0.99, 0.99, and 0.98, respectively, with overall prediction accuracies of 83.3%, 89.3%, 90.5%, 91.7%, and 89.3%. It indicated that the developed model was reliable and the machine learning algorithm combined with NIRS for origin identification was sufficiently feasible. OPLS-DA showed that Zingiberis Rhizoma from Sichuan(genuine producing areas) could be significantly distinguished from other regions, with good discriminative accuracy, suggesting that the NIRS established in this study combined with chemometrics can be used for the identification of Zingiberis Rhizoma from Sichuan. This study established a rapid and nondestructive identification and reliable data analysis method for origin identification of Zingiberis Rhizoma, which is expected to provide a new idea for the origin tracing of Chinese medicinal materials.
Algorithms ; Chemometrics ; China ; Ginger ; Least-Squares Analysis ; Plant Extracts ; Principal Component Analysis ; Support Vector Machine

Epimedii Folium is a well-known Chinese herbal medicine with the effect of nourishing kidney and strengthening Yang. Its main active ingredients are flavonoids. In this study, 60 samples of Epimedium sagittatum were collected for the determination of total flavonoids(TF) including the total amount of epimedin A, epimedin B, epimedin C, and icariin(abbreviated as ABCI) specified in the Chinese Pharmacopoeia as well as rhamnosylicariside Ⅱ and icariside Ⅱ. The calibration parameters of "first derivativemultiva-riate scattering correction in 1 900-650 cm~(-1) band(4-point smoothing)" and "first derivativestandard normal variable correction in 4 000-650 cm~(-1) full band(4-point smoothing)" were confirmed respectively. The quantitative model was established via Fourier infrared spectroscopy plus attenuated total reflection(FTIR-ATR) accessory combined with partial least squares(PLS) method and then used to predict the flavonoid content of 11 validation sets. The average prediction accuracy for ABCI in calibration set and validation set was 98.985% and 96.087%, respectively. The average prediction accuracy for TF in calibration set and validation set was 98.998% and 94.771%, respectively. These results indicated that FTIR-ATR combined with PLS model could be used for rapid prediction of flavonoid content in E. sagittatum, with the prediction accuracy above 94.7%. The establishment of this method provides a new solution for the detection of a large number of E. sagittatum samples.
Epimedium/chemistry* ; Flavonoids/chemistry* ; Plant Leaves ; Least-Squares Analysis ; Spectrophotometry, Infrared