Attention based residual network for medicinal fungi near infrared spectroscopy analysis

Lan Huang; Shuyu Guo; Ye Wang; Shang Wang; Qiubo Chu; Lu Li; Tian Bai; Lan Huang; Shuyu Guo; Ye Wang; Shang Wang; Qiubo Chu; Lu Li; Tian Bai

doi:10.3934/mbe.2019149

Mathematical Biosciences and Engineering

2019, Volume 16, Issue 4: 3003-3017. doi: 10.3934/mbe.2019149

Previous Article Next Article

Research article Special Issues

Attention based residual network for medicinal fungi near infrared spectroscopy analysis

1.
College of computer Science and Technology, Jilin University, Changchun 130012, China
2.
Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130012, China
3.
College of Life Sciences, Jilin University, Changchun 130012, China
4.
School of Software Engineering, Tongji University, Shanghai 201804, China

Received: 25 January 2019 Accepted: 25 March 2019 Published: 10 April 2019

As an effective technology, near infrared spectroscopy (NIRS) can be widely applied to analysis of active ingredients in medicinal fungi. Multiple regression methods are used to compute the relationship between spectral vectors and ingredient contents. In this paper, an autonomous feature extraction method by using attention based residual network (ABRN) to model original NIRS vectors is introduced. Attention module in ABRN is employed to enhance feature wave bands and to decay noise. Different from traditional NIRS analysis methods, ABRN does not require any preprocessing of artificial feature selections which rely on expert experience. The experiments test ABRN by analyzing original spectrums of medicinal fungi (Antrodia Camphorata and Matsutake), which are from 800 nm to 2500 nm, and predicting active ingredients within them. We compare ABRN with other popular NIRS analysis methods. The root mean square error of Antrodia Camphorata training set (RMSET) and validation set (RMSEV) are 0.0229 g·g^-1 and 0.0349 g·g^-1 for polysaccharide, and 0.0173 g·g^-1 and 0.0189 g·g^-1 for triterpene. The RMSET and RMSEV of Matsutake are 0.1343 g·g^-1 and 0.2472 g·g^-1 for polysaccharide, and 0.0328 g·g^-1 and 0.0445 g·g^-1 for ergosterol. The R² (coefficient of determination) of these four ingredients are 0.711, 0.753, 0.847 and 0.807. The results indicate that ABRN has better performance in autonomously extracting feature wave bands from original NIRS vectors, which can decrease the loss of tiny feature peaks.
- near infrared spectroscopy,
- medicinal fungi,
- residual network,
- attention mechanism,
- deep learning
Citation: Lan Huang, Shuyu Guo, Ye Wang, Shang Wang, Qiubo Chu, Lu Li, Tian Bai. Attention based residual network for medicinal fungi near infrared spectroscopy analysis[J]. Mathematical Biosciences and Engineering, 2019, 16(4): 3003-3017. doi: 10.3934/mbe.2019149

Related Papers:

Abstract

As an effective technology, near infrared spectroscopy (NIRS) can be widely applied to analysis of active ingredients in medicinal fungi. Multiple regression methods are used to compute the relationship between spectral vectors and ingredient contents. In this paper, an autonomous feature extraction method by using attention based residual network (ABRN) to model original NIRS vectors is introduced. Attention module in ABRN is employed to enhance feature wave bands and to decay noise. Different from traditional NIRS analysis methods, ABRN does not require any preprocessing of artificial feature selections which rely on expert experience. The experiments test ABRN by analyzing original spectrums of medicinal fungi (Antrodia Camphorata and Matsutake), which are from 800 nm to 2500 nm, and predicting active ingredients within them. We compare ABRN with other popular NIRS analysis methods. The root mean square error of Antrodia Camphorata training set (RMSET) and validation set (RMSEV) are 0.0229 g·g^-1 and 0.0349 g·g^-1 for polysaccharide, and 0.0173 g·g^-1 and 0.0189 g·g^-1 for triterpene. The RMSET and RMSEV of Matsutake are 0.1343 g·g^-1 and 0.2472 g·g^-1 for polysaccharide, and 0.0328 g·g^-1 and 0.0445 g·g^-1 for ergosterol. The R² (coefficient of determination) of these four ingredients are 0.711, 0.753, 0.847 and 0.807. The results indicate that ABRN has better performance in autonomously extracting feature wave bands from original NIRS vectors, which can decrease the loss of tiny feature peaks.

References

[1]	W. Guo, Y. Du, Y. Zhou, et al., At-line monitoring of key parameters of nisin fermentation by near infrared spectroscopy, chemometric modeling and model improvement, World J. Microbiol. Biotechnol., 28 (2012), 993–1002.
[2]	X. Zhang, W. Li, B. Yin, et al., Improvement of near infrared spectroscopic (NIRS) analysis of caffeine in roasted Arabica coffee by variable selection method of stability competitive adaptive reweighted sampling (SCARS), Spectrochim. Acta, Pt. A Mol. Biomol. Spectrosc., 114 (2013), 350–356.
[3]	Q. Meng, L. Teng, J. Lu, et al., Determination of methanol and ethanol synchronously in ternary mixture by NIRS and PLS regression, International Conference on Computational Science and Its Applications, (2005), 1040–1045.
[4]	A. D. A. Gomes, R. K. H. Galvao, M. C. U. Araujo, et al., The successive projections algorithm for interval selection in PLS, Microchem. J., 110 (2013), 202–208.
[5]	N. C. T. Mariani, G. H. D. A. Teixeira, K. M. G. Lima, et al., NIRS and iSPA-PLS for predicting total anthocyanin content in jaboticaba fruit, Food Chem., 174 (2015), 643–648.
[6]	X. Lai, J. Li, X. Gong, et al., Rapid simultaneous determination of andrographolides in andrographis paniculata by near-infrared spectroscopy, Anal. Lett., 51 (2018), 2745–2760.
[7]	Y. Wang, F. Li, M. Liu, et al., Rapid and nondestructive analysis of bacillus calmette–guerin polysaccharide nucleic acid injection by near-infrared spectroscopy with chemometrics, Anal. Lett., 51, 2375–2389.
[8]	G. Kim, S. Hong, A. Lee, et al., Moisture content measurement of broadleaf litters using near-infrared spectroscopy technique, Remote Sens., 9 (2017), 1212.
[9]	A. F. B. Magalhaes, G. H. D. A. Teixeira, A. C. H. Rios, et al., Prediction of meat quality traits in Nelore cattle by near-infrared reflectance spectroscopy, J. Anim. Sci., 96 (2018), 4229–4237.
[10]	T. Bai, L. Gong, Y. Wang, et al., A method for exploring implicit concept relatedness in biomedical knowledge network, BMC Bioinform., 17 (2016), 265.
[11]	Y. Wang, L. Huang, S. Guo, et al., A novel MEDLINE topic indexing method using image presentation, J. Vis. Commun. Image Represent, 58 (2019), 130–137.
[12]	L. Huang, Y. Wang, Y. Wang, et al., Gene-disease interaction retrieval from multiple sources: a network based method, Biomed. Res. Int., 2016 (2016), 3594517.
[13]	Y. Wang, H. Sun, W. Du, et al., Identification of essential proteins based on ranking edge-weights in protein-protein interaction networks, PLoS One, 9 (2014).
[14]	T. Bai, Y. Ge, C. Q. Yang, et al., BERST: An engine and tool for exploring biomedical entities and relationships, Chin. J. Electron., (2019), in press.
[15]	A. Guillen, F. G. D. Moral, L. J. Herrera, et al., Using near-infrared spectroscopy in the classification of white and iberian pork with neural networks, Neural Comput. Appl., 19 (2010), 465–470.
[16]	J. Lu, Y. B. Zhang, Z. Y. Zhang, et al., Application of wavelet transform-radial basis function neural network in NIRS for determination of rifampicin and isoniazide tablets, Spectrosc. Spectr. Anal., 28 (2008), 1264.
[17]	G. Xing, J. Cao, D. Wang, et al., Near infrared spectroscopic combined with partial least squares and radial basis function neural network to analyze paclitaxel concentration in rat plasma, Comb. Chem. High Throughput Screen, 18 (2015), 704–711.
[18]	J. Song, C. Li, G. Xing, et al., Study on analyzing active ingredient of marasmius and rosaceus via radial basis function neural network combining with near infrared spectroscopy, Acta Optica Sinica, 34 (2014), 320–325.
[19]	Y. Liu, C. Lu, Q. Meng, et al., Near infrared spectroscopy coupled with radial basis function neural network for at-line monitoring of Lactococcus lactis subsp. fermentation, Saudi J. Biol. Sci., 23 (2016).
[20]	L. Xie, X. Ye, D. Liu, et al., Application of principal component-radial basis function neural networks (PC-RBFNN) for the detection of water-adulterated bayberry juice by near-infrared spectroscopy, J. Zhejiang Univ. SCI. B, 9 (2008), 982–989.
[21]	J. T. Xue, Y. L. Shi, L. M. Ye, et al., Near-infrared spectroscopy for rapid and simultaneous determination of five main active components in rhubarb of different geographical origins and processing, Spectroc. Acta Pt. A-Molec. Biomolec. Spectr., 205 (2018), 419–427.
[22]	J. T. Xue, Y. F. Liu, L. M. Ye, et al., Rapid and simultaneous analysis of five alkaloids in four parts of Coptidis Rhizoma by near-infrared spectroscopy, Spectroc. Acta Pt. A-Molec. Biomolec. Spectr., 188 (2018), 611–618.
[23]	Y. Hui, G. Cheng, S. Yang, et al., Rapid detection of volatile oil in mentha haplocalyx by near-infrared spectroscopy and chemometrics, Pharmacogn. Mag., 13 (2017), 439–445.
[24]	Y. Dou, Y. Sun, Y. Ren, et al., Simultaneous non-destructive determination of two components of combined paracetamol and amantadine hydrochloride in tablets and powder by NIR spectroscopy and artificial neural networks, J. Pharm. Biomed. Anal., 37 (2005), 543–549.
[25]	Y. Roggo, P. Chalus, L. Maurer, et al., A review of near infrared spectroscopy and chemometrics in pharmaceutical technologies, J. Pharm. Biomed. Anal., 44 (2007), 683–700.
[26]	Y. Wang, S. Yang, J. Zhao, et al., Using machine learning to measure relatedness between genes: a multi-features model, Sci. Rep., 9 (2019), 4192.
[27]	S. Gutierrez, J. Tardaguila, J. Fernandeznovales, et al., Support vector machine and artificial neural network models for the classification of grapevine varieties using a portable NIR spectrophotometer, PLoS One, 10 (2015).
[28]	I. V. Kovalenko, G. R. Rippke and C. R. Hurburgh, Dimensionality reduction of near infrared spectral data using global and local implementations of principal component analysis for neural network calibrations, J. Near Infrared Spectrosc., 15 (2007), 21–28.
[29]	T. Bai, C. A. Kulikowski, L. Gong, et al., A global k-modes algorithm for clustering categorical data, Chin. J. Electron., 21 (2012), 460–465.
[30]	T. Bai, C. Wang, Y. Wang, et al., A novel deep learning method for extracting unspecific biomedical relation, Concurr. Comput. Pract. Exper., (2018), in press.
[31]	S. Liang, R. Zhang, D. Liang, et al., Multimodal 3D densenet for IDH genotype prediction in gliomas, Genes, 9 (2018), 382.
[32]	T. Liu, Z. Li, C. Yu, et al., NIRS feature extraction based on deep auto-encoder neural network, Infrared Phys. Technol., 87 (2017), 124–128.
[33]	H. Yang, B. Hu, X. Pan, et al., Deep belief network-based drug identification using near infrared spectroscopy, J. Innov. Opt. Health Sci., 10 (2017), 1630011.
[34]	E. J. Bjerrum, M. Glahder and T. Skov, Data augmentation of spectral data for convolutional neural network (CNN) based deep chemometrics, arXiv: Learning, (2017).
[35]	K. He, X. Zhang, S. Ren, et al., Deep residual learning for image recognition, Computer Vision and Pattern Recognition, (2016), 770–778.
[36]	D. P. Kingma and J. Ba, Adam: A method for stochastic optimization, International Conference on Learning Representations, (2015).
[37]	M. C. U. Araújo, T. C. B. Saldanha, R. K. H. Galvão, et al., The successive projections algorithm for variable selection in spectroscopic multicomponent analysis, Chemometrics Intell. Lab. Syst., 57 (2001), 65–73.

Reader Comments

your name:*

Email:*
© 2019 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)