Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Oct 25:14:1271320.
doi: 10.3389/fpls.2023.1271320. eCollection 2023.

Non-destructive prediction of isoflavone and starch by hyperspectral imaging and deep learning in Puerariae Thomsonii Radix

Huiqiang Hu  1   2 , Tingting Wang  1 , Yunpeng Wei  1 , Zhenyu Xu  1 , Shiyu Cao  3 , Ling Fu  3 , Huaxing Xu  1 , Xiaobo Mao  1   2 , Luqi Huang  1   2   4
Affiliations

Non-destructive prediction of isoflavone and starch by hyperspectral imaging and deep learning in Puerariae Thomsonii Radix

Huiqiang Hu et al. Front Plant Sci. .

Abstract

Accurate assessment of isoflavone and starch content in Puerariae Thomsonii Radix (PTR) is crucial for ensuring its quality. However, conventional measurement methods often suffer from time-consuming and labor-intensive procedures. In this study, we propose an innovative and efficient approach that harnesses hyperspectral imaging (HSI) technology and deep learning (DL) to predict the content of isoflavones (puerarin, puerarin apioside, daidzin, daidzein) and starch in PTR. Specifically, we develop a one-dimensional convolutional neural network (1DCNN) model and compare its predictive performance with traditional methods, including partial least squares regression (PLSR), support vector regression (SVR), and CatBoost. To optimize the prediction process, we employ various spectral preprocessing techniques and wavelength selection algorithms. Experimental results unequivocally demonstrate the superior performance of the DL model, achieving exceptional performance with mean coefficient of determination (R2) values surpassing 0.9 for all components. This research underscores the potential of integrating HSI technology with DL methods, thereby establishing the feasibility of HSI as an efficient and non-destructive tool for predicting the content of isoflavones and starch in PTR. Moreover, this methodology holds great promise for enhancing efficiency in quality control within the food industry.

Keywords: Puerariae Thomsonii Radix; deep learning; hyperspectral imaging; isoflavones and starch content; one-dimensional convolutional neural network.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
The CNN architecture utilized for predicting the levels of isoflavones and starch in PTR.
Figure 2
Figure 2
The mean, minimum, maximum and standard deviation spectral reflectance for PTR.
Figure 3
Figure 3
The regression results for the reference and predicted values of the four isoflavones and starch contents are depicted in (A-E), which illustrate the predictions of puerarin, puerarin apioside, daidzin, daidzein, and starch contents using the full wavelengths and conventional models.
Figure 4
Figure 4
The regression results for the reference and predicted values of the four isoflavones and starch contents are depicted in (A-E), which illustrate the predictions of puerarin, puerarin apioside, daidzin, daidzein, and starch contents using the full wavelengths and 1DCNN.
Figure 5
Figure 5
The specific locations of important wavelengths extracted by SPA, CARS, and UVE are presented in (A-E), showcasing the prediction of puerarin, puerarin apioside, daidzin, daidzein, and starch contents.
Figure 6
Figure 6
The regression results for the reference and predicted values of the four isoflavones and starch contents are depicted in (A-E), which illustrate the predictions of puerarin, puerarin apioside, daidzin, daidzein, and starch contents using the effective wavelengths and conventional models.
Figure 7
Figure 7
The regression results for the reference and predicted values of the four isoflavones and starch contents are displayed in (A-E), illustrating the prediction of puerarin, puerarin apioside, daidzin, daidzein, and starch contents using the effective wavelengths and 1DCNN.
See this image and copyright information in PMC

References

    1. Aredo V., Velásquez L., Siche R. (2017). Prediction of beef marbling using hyperspectral imaging (hsi) and partial least squares regression (plsr). Sci. Agropecuaria 8 (2), 169–174. doi: 10.17268/sci.agropecu.2017.02.09 - DOI
    1. Barbin D. F., ElMasry G., Sun D.-W., Allen P. (2013). Non-destructive determination of chemical composition in intact and minced pork using near-infrared hyperspectral imaging. Food Chem. 138 (2-3), 1162–1171. doi: 10.1016/j.foodchem.2012.11.120 - DOI - PubMed
    1. Cai Z., Huang Z., He M., Li C., Qi H., Peng J., et al. (2023). Identification of geographical origins of radix paeoniae alba using hyperspectral imaging with deep learning-based fusion approaches. Food Chem. 422, 136169. doi: 10.1016/j.foodchem.2023.136169 - DOI - PubMed
    1. Cai W., Li Y., Shao X. (2008). A variable selection method based on uninformative variable elimination for multivariate calibration of near-infrared spectra. Chemometrics Intelligent Lab. Syst. 90 (2), 188–194. doi: 10.1016/j.chemolab.2007.10.001 - DOI
    1. Chen Y., Cai W., Xu B. (2017). Phytochemical profiles of edible kudzu (pueraria thomsonii benth) grown in China as affected by thermal processing. J. Food Process. Preservation 41 (1), e12754.

LinkOut - more resources