Classification of Chinese Herbal Medicine Based on Improved LDA Algorithm Using Machine Olfaction

Luo, De Han; Shao, Ya Wen

doi:10.4028/www.scientific.net/amm.239-240.1532

Cited by 4 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Supervised pattern recognition algorithms such as linear discriminant analysis (LDA), partial least squares discriminant analysis (PLS-DA), soft independent modeling of class analogy (SIMCA), artificial neural networks (ANN), k-nearest neighbors (KNN), and least squares-support vector machine (LS-SVM) are often applied in HM authentication ( Huang et al, 2015 ). LDA can maximize the distance between classified samples or groups, enabling better differentiation among HM classes ( Luo and Shao, 2013 ). In the case of the number of samples is less than the number of measured variables (often referred to as ‘ill-conditioned), PLS-DA can be applied to improve the capability of the predictive models to classify the samples ( Razmovski-Naumovski et al, 2010 ).…”

Section: Fingerprint Data Analysis Using Chemometric Approachmentioning

confidence: 99%

Advances in Fingerprint Analysis for Standardization and Quality Control of Herbal Medicines

2022

View full text Add to dashboard Cite

Herbal drugs or herbal medicines (HMs) have a long-standing history as natural remedies for preventing and curing diseases. HMs have garnered greater interest during the past decades due to their broad, synergistic actions on the physiological systems and relatively lower incidence of adverse events, compared to synthetic drugs. However, assuring reproducible quality, efficacy, and safety from herbal drugs remains a challenging task. HMs typically consist of many constituents whose presence and quantity may vary among different sources of materials. Fingerprint analysis has emerged as a very useful technique to assess the quality of herbal drug materials and formulations for establishing standardized herbal products. Rather than using a single or two marker(s), fingerprinting techniques take great consideration of the complexity of herbal drugs by evaluating the whole chemical profile and extracting a common pattern to be set as a criterion for assessing the individual material or formulation. In this review, we described and assessed various fingerprinting techniques reported to date, which are applicable to the standardization and quality control of HMs. We also evaluated the application of multivariate data analysis or chemometrics in assisting the analysis of the complex datasets from the determination of HMs. To ensure that these methods yield reliable results, we reviewed the validation status of the methods and provided perspectives on those. Finally, we concluded by highlighting major accomplishments and presenting a gap analysis between the existing techniques and what is needed to continue moving forward.

show abstract

Section: Fingerprint Data Analysis Using Chemometric Approachmentioning

confidence: 99%

Advances in Fingerprint Analysis for Standardization and Quality Control of Herbal Medicines

2022

View full text Add to dashboard Cite

show abstract

“…Supervised pattern recognition in HMs' authentication was commonly implemented in generating the classification model according to experimental data by assigning unknown samples to a previously labelled sample class according to the pattern properties of chemical measurements [39]. LDA can be applied for maximizing the classification distance between several classes of HMs samples [40]. In the case of 'ill conditioned', when the number of samples are less than the number of the measured variables, as commonly found in chemistry data, the PLS-DA may improve the classification capability of the generated predictive model [41].…”

Section: Chemometricsmentioning

confidence: 99%

Metabolite Fingerprinting Based on 1H-NMR Spectroscopy and Liquid Chromatography for the Authentication of Herbal Products

et al. 2022

View full text Add to dashboard Cite

Herbal medicines (HMs) are regarded as one of the traditional medicines in health care to prevent and treat some diseases. Some herbal components such as turmeric and ginger are used as HMs, therefore the identification and confirmation of herbal use are very necessary. In addition, the adulteration practice, mainly motivated to gain economical profits, may occur by substituting the high price of HMs with lower-priced ones or by addition of certain chemical constituents known as Bahan Kimia Obat (chemical drug ingredients) in Indonesia. Some analytical methods based on spectroscopic and chromatographic methods are developed for the authenticity and confirmation of the HMs used. Some approaches are explored during HMs authentication including single-component analysis, fingerprinting profiles, and metabolomics studies. The absence of reference standards for certain chemical markers has led to exploring the fingerprinting approach as a tool for the authentication of HMs. During fingerprinting-based spectroscopic and chromatographic methods, the data obtained were big, therefore the use of chemometrics is a must. This review highlights the application of fingerprinting profiles using variables of spectral and chromatogram data for authentication in HMs. Indeed, some chemometrics techniques, mainly pattern recognition either unsupervised or supervised, were applied for this purpose.

show abstract

“…LDA is a supervised method for classification by finding decision surfaces and calculating the signed orthogonal distance of data points. It has been used in the identification of Chinese herbal medicines [ 18 ]. SVM is a kind of regularization in which the aim is to find the maximum margin between classes.…”

Section: Introductionmentioning

confidence: 99%

An Odor Labeling Convolutional Encoder–Decoder for Odor Sensing in Machine Olfaction

Wen

et al. 2021

Sensors

View full text Add to dashboard Cite

Deep learning methods have been widely applied to visual and acoustic technology. In this paper, we propose an odor labeling convolutional encoder–decoder (OLCE) for odor identification in machine olfaction. OLCE composes a convolutional neural network encoder and decoder where the encoder output is constrained to odor labels. An electronic nose was used for the data collection of gas responses followed by a normative experimental procedure. Several evaluation indexes were calculated to evaluate the algorithm effectiveness: accuracy 92.57%, precision 92.29%, recall rate 92.06%, F1-Score 91.96%, and Kappa coefficient 90.76%. We also compared the model with some algorithms used in machine olfaction. The comparison result demonstrated that OLCE had the best performance among these algorithms.

show abstract

Classification of Chinese Herbal Medicine Based on Improved LDA Algorithm Using Machine Olfaction

Cited by 4 publications

References 10 publications

Advances in Fingerprint Analysis for Standardization and Quality Control of Herbal Medicines

Advances in Fingerprint Analysis for Standardization and Quality Control of Herbal Medicines

Metabolite Fingerprinting Based on 1H-NMR Spectroscopy and Liquid Chromatography for the Authentication of Herbal Products

An Odor Labeling Convolutional Encoder–Decoder for Odor Sensing in Machine Olfaction

Contact Info

Product

Resources

About