Combination of Modified Optical Path Length Estimation and Correction and Moving Window Partial Least Squares to Waveband Selection for the Fourier Transform Near-Infrared Determination of Pectin in Shaddock Peel

Chen, Huazhou; Tang, Guoqiang; Song, Qi-Qing; Ai, Wu

doi:10.1080/00032719.2013.784912

Cited by 14 publications

(10 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It has been proved concerning the ability of informative spectral selection in the regression developments by our studies and others. 17,[24][25][26] As for MWPLS-DA, 27 an algorithm of MWPLSR was employed in the calculation procedure prior the development of the PLS model for classification. A schematic diagram of the calculation procedure is given in Fig.…”

Section: Moving Window Partial Least Squares-discrimination Analysis mentioning

confidence: 99%

Rapid Classification of Turmeric Based on DNA Fingerprint by Near-Infrared Spectroscopy Combined with Moving Window Partial Least Squares-Discrimination Analysis

2017

View full text Add to dashboard Cite

In this research, near-infrared (NIR) spectroscopy in combination with moving window partial least squares-discrimination analysis (MWPLS-DA) was utilized to discriminate the variety of turmeric based on DNA markers, which correlated to the quantity of curcuminoid. Curcuminoid was used as a marker compound in variety identification due to the most pharmacological properties of turmeric possessed from it. MWPLS-DA optimized informative NIR spectral regions for the fitting and prediction to {-1/1}-coded turmeric varieties, indicating variables in the development of latent variables in discrimination analysis. Consequently, MWPLS-DA benefited in the selection of combined informative NIR spectral regions of 1100 -1260, 1300 -1500 and 1880 -2500 nm for classification modeling of turmeric variety with 148 calibration samples, and yielded the results better than that obtained from a partial least squares-discrimination analysis (PLS-DA) model built by using the whole NIR spectral region. An effective and rapid strategy of using NIR in combination with MWPLS-DA provided the best variety identification results of 100% in both specificity and total accuracy for 48 test samples.

show abstract

Section: Moving Window Partial Least Squares-discrimination Analysis mentioning

confidence: 99%

Rapid Classification of Turmeric Based on DNA Fingerprint by Near-Infrared Spectroscopy Combined with Moving Window Partial Least Squares-Discrimination Analysis

2017

View full text Add to dashboard Cite

show abstract

“…The PLS latent valuable is a major parameter that corresponds to the number of integrated variables in spectral responses of samples. The selection of a suitable number of latent valuables is both necessary and difficult (Chen et al , 2013(Chen et al , 2014. Thus, in summary, an MWPLS model is established in consideration of three tunable parameters, of which two are used to determine the informative window; the other one is the optimal PLS modeling latent valuable (hereafter, we denote PLS latent valuable as LV).…”

Section: The Methods Of Parametric Mwplsmentioning

confidence: 99%

“…For multiplicative effect correction, MOPLEC is a simple and effective approach preprocessing method with an improvement of robustness (Jin et al 2012;Wang et al 2011;Chen et al 2013).…”

Section: The Preprocessing Methods Of Moplecmentioning

confidence: 99%

“…The noise elimination should be considered as another important part for the preprocessing on the raw measured spectra. As for this, a useful data preprocessing method realizes the full potential for quantitative spectroscopic analysis of complex systems by using a strategy of duel calibration, which is called modified optical path length estimation and correction (MOPLEC) (Jin et al 2012;Wang et al 2011;Chen et al 2013). The MOPLEC method is expected to be used in the noise elimination for the spectra of Shatian pomelo samples.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Rapid Detection of Surface Color of Shatian Pomelo Using Vis-NIR Spectrometry for the Identification of Maturity

Chen

Tang

et al. 2015

Food Anal. Methods

View full text Add to dashboard Cite

Shatian pomelo is one kind of high-quality fruits grown in south China. Modern agricultural planting technology for Shatian pomelo closely relates to the rapid identification of its maturity. As surface color reflects the maturity of fruits, quantitative determination on the color indices of L*, a*, and b* is significant for maturity judgment. Chemoinformatic models were applied for quantitative detection of L*, a*, and b* for Shatian pomelo, by using visible and near-infrared (Vis-NIR) spectrometry. To enhance the analytical accuracy and model stability, we proposed the combined use of moving window partial least squares (MWPLS) and modified optical path length estimation and correction (MOPLEC) to search for high-signal wavelength combinations, in fully optimizing the modeling parameters. Results indicated that the chemoinformatics that we proposed performed well for nondestructive evaluations of the maturity of Shatian pomelo. This technology is expected to be utilized for quality evaluations of some other agricultural and planting fruits.

show abstract

“…Partial least squares (PLS) regression is the most routinely utilized in spectroscopic analytical applications, but a single PLS‐based classification method is hard to perform best for all different datasets in clinical diagnosis of blood plasma/serum because PLS models are always coupled with data colinearity. Under this situation, some other chemometric methods have been also proposed …”

Section: Introductionmentioning

confidence: 99%

Use of random forest in FTIR analysis of LDL cholesterol and tri‐glycerides for hyperlipidemia

Chen

Tang

et al. 2015

Biotechnology Progress

Self Cite

View full text Add to dashboard Cite

A quantitative determination method for the diagnosis of hyperlipidemia was developed using Fourier transform infrared (FTIR) spectroscopy. Random forest (RF) was demonstrated as a potential multivariate algorithm for the FTIR analysis of low-density lipoprotein cholesterol (LDL-C) and tri-glycerides (TG) in human serum samples. The informative wavebands for LDL-C and TG were selected based on the Gini importance. The selected wavebands were mainly within the fingerprint region. The RF modeling results were better than those derived using PLS in validation process, because the chance for over-fitting was possibly eliminated in RF algorithm. ARF also demonstrated favorable results in the test process. The prospective model exhibited a higher than 90% true prediction in negative/positive properties for male and female samples. These clinical statistical results indicated the optimization of RF algorithm performed accurately in the FTIR determination of LDL-C and TG. RF is evaluated as a promising tool for diagnosing and controlling hyperlipidemia in populations. The parameter optimization methodology is useful in the improving model accuracy using FTIR spectroscopic technology.

show abstract

Combination of Modified Optical Path Length Estimation and Correction and Moving Window Partial Least Squares to Waveband Selection for the Fourier Transform Near-Infrared Determination of Pectin in Shaddock Peel

Cited by 14 publications

References 20 publications

Rapid Classification of Turmeric Based on DNA Fingerprint by Near-Infrared Spectroscopy Combined with Moving Window Partial Least Squares-Discrimination Analysis

Rapid Classification of Turmeric Based on DNA Fingerprint by Near-Infrared Spectroscopy Combined with Moving Window Partial Least Squares-Discrimination Analysis

Rapid Detection of Surface Color of Shatian Pomelo Using Vis-NIR Spectrometry for the Identification of Maturity

Use of random forest in FTIR analysis of LDL cholesterol and tri‐glycerides for hyperlipidemia

Contact Info

Product

Resources

About