Two dimensional correlation spectroscopy combined with ResNet: Efficient method to identify bolete species compared to traditional machine learning

Application of ATR-FTIR and FT-NIR spectroscopy coupled with chemometrics for species identification and quality prediction of boletes

The taste and aroma of edible mushrooms, which is a criterion of judgment for consumer purchases, are influenced by amino acids and their metabolites. Sixty-eight amino acids and their metabolites were identified using liquid chromatography mass spectrometry (LC-MS), and 16 critical marker components were screened. The chemical composition of different species of boletes was characterized by two-dimensional correlation spectroscopy (2DCOS) to determine the sequence of molecular vibrations or group changes. Identification of boletes species based on partial least squares discrimination (PLS-DA) combined with Fourier transform near-infrared spectroscopy (FT-NIR) and Fourier transform infrared spectroscopy (ATR-FTIR), residual convolutional neural network (ResNet) combined with three-dimensional correlation spectroscopy (3DCOS) was performed with 100% accuracy. Partial least squares regression (PLSR) analysis showed that FT-NIR and ATR-FTIR spectra were highly correlated with the amino acids and their metabolites detected by LC-MS. All models had achieved an R2p of 0.911 and an RPD >3.0. The results show that FT-NIR and ATR-FTIR spectroscopy in combination with chemometrics methods can be used for rapid species identification and estimation of amino acids and their metabolites content in boletes. This study provides new techniques and ideas for the authenticity of species information and the quality assessment of boletes.

Identification of geographical origins of Gastrodia elata Blume based on multisource data fusion

INTRODUCTION

Identifying the geographical origin of Gastrodia elata Blume contributes to the scientific and rational utilization of medicinal materials. In this study, infrared spectroscopy was combined with machine learning algorithms to distinguish the origin of G. elata BI.

OBJECTIVE

Realization of rapid and accurate identification of the origin of G. elata BI.

MATERIALS AND METHODS

Attenuated total reflection Fourier transform infrared (ATR-FTIR) spectra and Fourier transform near-infrared (FT-NIR) spectra were collected for 306 samples of G. elata BI.

SAMPLES

Firstly, a support vector machine (SVM) model was established based on the single-spectrum and the full-spectrum fusion data. To investigate whether feature-level fusion strategy can enhance the model's performance, the sequential and orthogonalized partial least squares discriminant analysis (SO-PLS-DA) model was established to extract and combine two types of spectral features. Next, six algorithms were employed to extract feature variables, SVM model was established based on the feature-level fusion data. To avoid complicated preprocessing and feature extraction processes, a residual convolutional neural network (ResNet) model was established after converting the raw spectral data into spectral images.

RESULTS

The accuracy of the feature-level fusion model is better as compared to the single-spectrum model and the fusion model with full-spectrum, and SO-PLS-DA is simpler than feature-level fusion based on the SVM model. The ResNet model performs well in classification but requires more data to enhance its generalization capability and training effectiveness.

CONCLUSION

Sequential and orthogonalized data fusion approaches and ResNet models are powerful solutions for identifying the geographic origin of G. elata BI.

Novel Developments and Progress in Two-Dimensional Correlation Spectroscopy (2D-COS)

This first of the two-part series of the comprehensive survey review on the progress of the two-dimensional correlation spectroscopy (2D-COS) field during the period 2021-2022, covers books, reviews, tutorials, novel concepts and theories, and patent applications that appeared in the last two years, as well as some inappropriate use or citations of 2D-COS. The overall trend clearly shows that 2D-COS is continually growing and evolving with notable new developments. The technique is well recognized as a powerful analytical tool that provides deep insights into systems in many science fields.

Diverse Applications of Two-Dimensional Correlation Spectroscopy (2D-COS)

This second of the two-part series of a comprehensive survey review provides the diverse applications of two-dimensional correlation spectroscopy (2D-COS) covering different probes, perturbations, and systems in the last two years. Infrared spectroscopy has maintained its top popularity in 2D-COS over the past two years. Fluorescence spectroscopy is the second most frequently used analytical method, which has been heavily applied to the analysis of heavy metal binding, environmental, and solution systems. Various other analytical methods including laser-induced breakdown spectroscopy, dynamic mechanical analysis, differential scanning calorimetry, capillary electrophoresis, seismologic, and so on, have also been reported. In the last two years, concentration, composition, and pH are the main effects of perturbation used in the 2D-COS fields, as well as temperature. Environmental science is especially heavily studied using 2D-COS. This comprehensive survey review shows that 2D-COS undergoes continuous evolution and growth, marked by novel developments and successful applications across diverse scientific fields.

A rapid method for identification of Lanxangia tsaoko origin and fruit shape: FT-NIR combined with chemometrics and image recognition

Lanxangia tsaoko's accurate classifications of different origins and fruit shapes are significant for research in L. tsaoko difference between origin and species as well as for variety breeding, cultivation, and market management. In this work, Fourier transform-near infrared (FT-NIR) spectroscopy was transformed into two-dimensional and three-dimensional correlation spectroscopies to further investigate the spectral characteristics of L. tsaoko. Before building the classification model, the raw FT-NIR spectra were preprocessed using multiplicative scatter correction and second derivative, whereas principal component analysis, successive projections algorithm, and competitive adaptive reweighted sampling were used for spectral feature variable extraction. Then combined with partial least squares-discriminant analysis (PLS-DA), support vector machine (SVM), decision tree, and residual network (ResNet) models for origin and fruit shape discriminated in L. tsaoko. The PLS-DA and SVM models can achieve 100% classification in origin classification, but what is difficult to avoid is the complex process of model optimization. The ResNet image recognition model classifies the origin and shape of L. tsaoko with 100% accuracy, and without the need for complex preprocessing and feature extraction, the model facilitates the realization of fast, accurate, and efficient identification.

An integrated chemical characterization based on FT-NIR, and GC-MS for the comparative metabolite profiling of 3 species of the genus Amomum

BACKGROUND

The fruits and seeds of genus Amomum are well-known as medicinal plants and edible spices, and are used in countries such as China, India and Vietnam to treat malaria, gastrointestinal disorders and indigestion. The morphological differences between different species are relatively small, and technical characterization and identification techniques are needed.

RESULTS

Fourier transform near infrared spectroscopy (FT-NIR) and gas chromatography-mass spectrometry (GC-MS), combined with principal component analysis and two-dimensional correlation analysis were used to characterize the chemical differences of Amomum tsao-ko, Amomum koenigii, and Amomum paratsaoko. The targets and pathways for the treatment of diabetes mellitus in three species were predicted using network pharmacology and screened for the corresponding pharmacodynamic components as potential quality markers. The results of "component-target-pathway" network showed that (+)-Nerolidol, 2-Nonanol, α-Terpineol, α-Pinene, 2-Nonanone had high degree values and may be the main active components. Partial least squares-discriminant analysis (PLS-DA) was further used to select for differential metabolites and was identified as a potential quality marker, 11 in total. PLS-DA and residual network (ResNet) classification models were developed for the identification of 3 species of the genus Amomum, ResNet model is more suitable for the identification study of large volume samples.

SIGNIFICANCE

This study characterizes the differences between the three species in a visual way and also provides a reliable technique for their identification, while demonstrating the ability of FT-NIR spectroscopy for fast, easy and accurate species identification. The results of this study lay the foundation for quality evaluation studies of genus Amomum and provide new ideas for the development of new drugs for the treatment of diabetes mellitus.

Building deep learning and traditional chemometric models based on Fourier transform mid-infrared spectroscopy: Identification of wild and cultivated Gastrodia elata

To identify wild and cultivated Gastrodia elata quickly and accurately, this study is the first to apply three-dimensional correlation spectroscopy (3DCOS) images combined with deep learning models to the identification of G. elata. The spectral data used for model building do not require any preprocessing, and the spectral data are converted into three-dimensional spectral images for model building. For large sample studies, the time cost is minimized. In addition, a partial least squares discriminant analysis (PLS-DA) model and a support vector machine (SVM) model are built for comparison with the deep learning model. The overall effect of the deep learning model is significantly better than that of the traditional chemometric models. The results show that the model achieves 100% accuracy in the training set, test set, and external validation set of the model built after 46 iterations without preprocessing the original spectral data. The sensitivity, specificity, and the effectiveness of the model are all 1. The results concluded that the deep learning model is more effective than the traditional chemometric model and has greater potential for application in the identification of wild and cultivated G. elata.

Identification of Gentian-Related Species Based on Two-Dimensional Correlation Spectroscopy (2D-COS) Combined with Residual Neural Network (ResNet)

Gentian is a traditional Chinese herb with heat-clearing, damp-drying, inflammation-alleviating and digestion-promoting effects, which is widely used in clinical practice. However, there are many species of gentian. According to the pharmacopoeia, Gentiana manshurica Kitag, Gentiana scabra Bge, Gentiana triflora Pall and Gentianarigescens Franch are included. Therefore, accurately identifying the species of gentian is important in clinical use. In recent years, with the advantages of low cost, convenience, fast analysis and high sensitivity, infrared spectroscopy (IR) has been extensively used in herbal identification. Unlike one-dimensional spectroscopy, a two-dimensional correlation spectrum (2D-COS) can improve the resolution of the spectrum and better highlight the details that are difficult to detect. In addition, the residual neural network (ResNet) is an important breakthrough in convolutional neural networks (CNNs) for significant advantages related to image recognition. Herein, we propose a new method for identifying gentian-related species using 2D-COS combined with ResNet. A total of 173 gentian samples from seven different species are collected in this study. In order to eliminate a large amount of redundant information and improve the efficiency of machine learning, the extracted feature band method was used to optimize the model. Four feature bands were selected from the infrared spectrum, namely 3500-3000 cm^-1, 3000-2750 cm^-1, 1750-1100 cm^-1 and 1100-400 cm^-1, respectively. The one-dimensional spectral data were converted into synchronous 2D-COS images, asynchronous 2D-COS images, and integrative 2D-COS images using Matlab (R2022a). The identification strategy for these three 2D-COS images was based on ResNet, which analyzes 2D-COS images based on single feature bands and full bands as well as fused feature bands. According to the results, (1) compared with the other two 2D-COS images, synchronous 2D-COS images are more suitable for the ResNet model, and (2) after extracting a single feature band 1750-1100 cm^-1 to optimize ResNet, the model has the best convergence performance, the accuracy of training, test and external validation is 1 and the loss value is only 0.155. In summary, 2D-COS combined with ResNet is an effective and accurate method to identify gentian-related species.

Rapid and Accurate Authentication of Porcini Mushroom Species Using Fourier Transform Near-Infrared Spectra Combined with Machine Learning and Chemometrics

Porcini mushrooms have high nutritional value and great potential, but different species are easily confused, so it is essential to identify them rapidly and precisely. The diversity of nutrients in stipe and cap will lead to differences in spectral information. In this research, Fourier transform near-infrared (FT-NIR) spectral information about imparity species of porcini mushroom stipe and cap was collected and combined into four data matrices. FT-NIR spectra of four data sets were combined with chemometric methods and machine learning for accurate evaluation and identification of different porcini mushroom species. From the results: (1) improved visualization level of t-distributed stochastic neighbor embedding (t-SNE) results after the second derivative preprocessing compared with raw spectra; (2) after using multiple pretreatment combinations to process the four data matrices, the model accuracies based on support vector machine and partial least-square discriminant analysis (PLS-DA) under the best preprocessing method were 98.73-99.04% and 98.73-99.68%, respectively; (3) by comparing the modeling results of FT-NIR spectra with different data matrices, it was found that the PLS-DA model based on low-level data fusion has the highest accuracy (99.68%), but residual neural network (ResNet) model based on the stipe, cap, and average spectral data matrix worked better (100% accuracy). The above results suggest that distinct models should be selected for dissimilar spectral data matrices of porcini mushrooms. Additionally, FT-NIR spectra have the advantages of being nondevastate and fast; this method is expected to be a promising analytical tool in food safety control.

Application of spectral image processing with different dimensions combined with large-screen visualization in the identification of boletes species

Boletes are favored by consumers because of their unique flavor, rich nutrition and delicious taste. However, the different nutritional values of each species lead to obvious price differences, so shoddy products appear on the market, which affects food safety. The aim of this study was to find a rapid and effective method for boletes species identification. In this paper, 1,707 samples of eight boletes species were selected as the research objects. The original Mid-Infrared (MIR) spectroscopy data were adopted for support vector machine (SVM) modeling. The 11,949 spectral images belong to seven data sets such as two-dimensional correlation spectroscopy (2DCOS) and three-dimensional correlation spectroscopy (3DCOS) were used to carry out Alexnet and Residual network (Resnet) modeling, thus we established 15 models for the identification of boletes species. The results show that the SVM method needs to process complex feature data, the time cost is more than 11 times of other models, and the accuracy is not high enough, so it is not recommended to be used in data processing with large sample size. From the perspective of datasets, synchronous 2DCOS and synchronous 3DCOS have the best modeling results, while one-dimensional (1D) MIR Spectrum dataset has the worst modeling results. After comprehensive analysis, the modeling effect of Resnet on the synchronous 2DCOS dataset is the best. Moreover, we use large-screen visualization technology to visually display the sample information of this research and obtain their distribution rules in terms of species and geographical location. This research shows that deep learning combined with 2DCOS and 3DCOS spectral images can effectively and accurately identify boletes species, which provides a reference for the identification of other fields, such as food and Chinese herbal medicine.

An Intelligent Waste-Sorting and Recycling Device Based on Improved EfficientNet

The main source of urban waste is the daily life activities of residents, and the waste sorting of residents' waste is important for promoting economic recycling, reducing labor costs, and protecting the environment. However, most residents are unable to make accurate judgments about the categories of household waste, which severely limits the efficiency of waste sorting. We have designed an intelligent waste bin that enables automatic waste sorting and recycling, avoiding the extensive knowledge required for waste sorting. To ensure that the waste-classification model is high accuracy and works in real time, GECM-EfficientNet is proposed based on EfficientNet by streamlining the mobile inverted bottleneck convolution (MBConv) module, introducing the efficient channel attention (ECA) module and coordinate attention (CA) module, and transfer learning. The accuracy of GECM-EfficientNet reaches 94.54% and 94.23% on the self-built household waste dataset and TrashNet dataset, with parameters of only 1.23 M. The time of one recognition on the intelligent waste bin is only 146 ms, which satisfies the real-time classification requirement. Our method improves the computational efficiency of the waste-classification model and simplifies the hardware requirements, which contributes to the residents' waste classification based on intelligent devices.

Anti-Nutritional Factors and Protein Dispersibility Index as Principal Quality Indicators for Soybean Meal in Diet of Nile Tilapia (Oreochromis niloticus GIFT), a Meta-Analysis

Soybean meal (SBM) is the most important plant protein source in animal feed. This study investigated the characteristics of different SBMs, produced by soybeans from America and Brazil (SBM-A and SBM-B) in 2017−2021 under the same controlled conditions. The effects of different SBMs on the growth performance of Nile tilapia (Oreochromis niloticus, GIFT) and apparent digestibility coefficients (ADCs) of nutrients and energy were studied. The results showed that protein dispersibility index (PDI), urease activity (UA), glycinin and fiber were the four primary key indicators for distinguishing the characteristics of the tested SBMs. The meta-analysis results suggested that UA, glycinin, and fiber showed a negative effect on the survival rate (SR) and weight gain rate (WGR) of the Nile tilapia, whereas β-conglycinin, PDI, and nitrogen solubility index (NSI) had a positive effect on the SR and WGR of the fish. The ADCs of dry matter, the gross energy, phosphorus, crude protein, valine (Val), lysine (Lys), histidine (His), serine (Ser), and glutamate (Glu) of the Diet-A group (SBM-A inclusion) were significantly higher than those in the Diet-B group (SBM-B inclusion) (p < 0.05). However, no significant difference was found in ADCs of macro-nutrients between the two SBMs (p > 0.05). Overall, PDI, UA, glycinin, and fiber were the main indicators reflecting the characteristics of the tested SBMs, and UA, glycinin, β-conglycinin, and PDI had the greatest impact on the growth performance of Nile tilapia in this study. PDI was a more sensitive indicator than NSI for representing the protein quality of SBM.