Cargando…

Machine learning based on structural and FTIR spectroscopic datasets for seed autoclassification

A single feature set is often unable to effectively classify complex biological samples due to their similar morphology and sizes. This paper proposes a protocol for the fast identification of seed medicinal materials based on micro-structural and infrared spectroscopic characteristics. Three differ...

Descripción completa

Detalles Bibliográficos
Autores principales: Wang, Hanqiu, Rehmetulla, Aybek, Guo, Shanshan, Kong, Xin, Lü, Zhiwei, Guan, Yu, Xu, Cong, Sulaiman, Kaiser, Wei, Gongxiang, Liu, Huiqiang
Formato: Online Artículo Texto
Lenguaje:English
Publicado: The Royal Society of Chemistry 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9004588/
https://www.ncbi.nlm.nih.gov/pubmed/35425064
http://dx.doi.org/10.1039/d2ra00239f
_version_ 1784686301084123136
author Wang, Hanqiu
Rehmetulla, Aybek
Guo, Shanshan
Kong, Xin
Lü, Zhiwei
Guan, Yu
Xu, Cong
Sulaiman, Kaiser
Wei, Gongxiang
Liu, Huiqiang
author_facet Wang, Hanqiu
Rehmetulla, Aybek
Guo, Shanshan
Kong, Xin
Lü, Zhiwei
Guan, Yu
Xu, Cong
Sulaiman, Kaiser
Wei, Gongxiang
Liu, Huiqiang
author_sort Wang, Hanqiu
collection PubMed
description A single feature set is often unable to effectively classify complex biological samples due to their similar morphology and sizes. This paper proposes a protocol for the fast identification of seed medicinal materials based on micro-structural and infrared spectroscopic characteristics. Three different feature datasets, namely micro-CT, FTIR, and mixed datasets, were established via principal component analysis (PCA) and competitive adaptive reweighted sampling (CARS) and then used to train a back-propagation neural network. The mixed dataset consists of 34-dimensional micro-CT eigenvalues and 13-dimensional FTIR eigenvalues, optimized by PCA and CARS processing and then used to train a BP neural network. The results showed that the classification accuracy reached 89.5% for the micro-CT dataset and 93.3% for the FTIR dataset, and the classification accuracy of the mixed dataset achieved 99.2%, much higher than those of the traditional single feature datasets. This study provides a new protocol for multi-dimensional characteristic architecture with excellent performance for the classification and identification of Chinese medicinal materials.
format Online
Article
Text
id pubmed-9004588
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher The Royal Society of Chemistry
record_format MEDLINE/PubMed
spelling pubmed-90045882022-04-13 Machine learning based on structural and FTIR spectroscopic datasets for seed autoclassification Wang, Hanqiu Rehmetulla, Aybek Guo, Shanshan Kong, Xin Lü, Zhiwei Guan, Yu Xu, Cong Sulaiman, Kaiser Wei, Gongxiang Liu, Huiqiang RSC Adv Chemistry A single feature set is often unable to effectively classify complex biological samples due to their similar morphology and sizes. This paper proposes a protocol for the fast identification of seed medicinal materials based on micro-structural and infrared spectroscopic characteristics. Three different feature datasets, namely micro-CT, FTIR, and mixed datasets, were established via principal component analysis (PCA) and competitive adaptive reweighted sampling (CARS) and then used to train a back-propagation neural network. The mixed dataset consists of 34-dimensional micro-CT eigenvalues and 13-dimensional FTIR eigenvalues, optimized by PCA and CARS processing and then used to train a BP neural network. The results showed that the classification accuracy reached 89.5% for the micro-CT dataset and 93.3% for the FTIR dataset, and the classification accuracy of the mixed dataset achieved 99.2%, much higher than those of the traditional single feature datasets. This study provides a new protocol for multi-dimensional characteristic architecture with excellent performance for the classification and identification of Chinese medicinal materials. The Royal Society of Chemistry 2022-04-12 /pmc/articles/PMC9004588/ /pubmed/35425064 http://dx.doi.org/10.1039/d2ra00239f Text en This journal is © The Royal Society of Chemistry https://creativecommons.org/licenses/by-nc/3.0/
spellingShingle Chemistry
Wang, Hanqiu
Rehmetulla, Aybek
Guo, Shanshan
Kong, Xin
Lü, Zhiwei
Guan, Yu
Xu, Cong
Sulaiman, Kaiser
Wei, Gongxiang
Liu, Huiqiang
Machine learning based on structural and FTIR spectroscopic datasets for seed autoclassification
title Machine learning based on structural and FTIR spectroscopic datasets for seed autoclassification
title_full Machine learning based on structural and FTIR spectroscopic datasets for seed autoclassification
title_fullStr Machine learning based on structural and FTIR spectroscopic datasets for seed autoclassification
title_full_unstemmed Machine learning based on structural and FTIR spectroscopic datasets for seed autoclassification
title_short Machine learning based on structural and FTIR spectroscopic datasets for seed autoclassification
title_sort machine learning based on structural and ftir spectroscopic datasets for seed autoclassification
topic Chemistry
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9004588/
https://www.ncbi.nlm.nih.gov/pubmed/35425064
http://dx.doi.org/10.1039/d2ra00239f
work_keys_str_mv AT wanghanqiu machinelearningbasedonstructuralandftirspectroscopicdatasetsforseedautoclassification
AT rehmetullaaybek machinelearningbasedonstructuralandftirspectroscopicdatasetsforseedautoclassification
AT guoshanshan machinelearningbasedonstructuralandftirspectroscopicdatasetsforseedautoclassification
AT kongxin machinelearningbasedonstructuralandftirspectroscopicdatasetsforseedautoclassification
AT luzhiwei machinelearningbasedonstructuralandftirspectroscopicdatasetsforseedautoclassification
AT guanyu machinelearningbasedonstructuralandftirspectroscopicdatasetsforseedautoclassification
AT xucong machinelearningbasedonstructuralandftirspectroscopicdatasetsforseedautoclassification
AT sulaimankaiser machinelearningbasedonstructuralandftirspectroscopicdatasetsforseedautoclassification
AT weigongxiang machinelearningbasedonstructuralandftirspectroscopicdatasetsforseedautoclassification
AT liuhuiqiang machinelearningbasedonstructuralandftirspectroscopicdatasetsforseedautoclassification