Machine learning classification of origins and varieties of Tetrastigma hemsleyanum using a dual-mode microscopic hyperspectral imager

Nondestructive detection of saponin content in Panax notoginseng powder based on hyperspectral imaging

This study investigated the feasibility of using hyperspectral imaging (HSI) technique to detect the saponin content in Panax notoginseng (PN) powder. The reflectance hyperspectral images of PN powder samples were collected in the spectral range of 400.6-999.9 nm. Savitzky-golay (SG) smoothing combined with detrending correction was utilized to preprocess the original spectral data. Two model population analysis (MPA) based methods, namely bootstrapping soft shrinkage (BOSS) and iteratively retains informative variables (IRIV) were employed to extract feature wavelengths from the full spectra. A generalized normal distribution optimization based extreme learning machine (GNDO-ELM) model was proposed to establish calibration model between spectra and saponin content, and compared with existing methods (GA-ELM, PSO-ELM and SSA-ELM). The result showed that the IRIV-GNDO-ELM model gave the best performance, with coefficient of determination for prediction (R2P) of 0.953 and root mean square error for prediction (RMSEP) of 0.115%. Therefore, it is possible to determine the saponin content of PN powder by using HSI technique.

An aberration-free line scan confocal Raman imager and type classification and distribution detection of microplastics

An aberration-free line scanning confocal Raman imager (named AFLSCRI) is developed to achieve rapid Raman imaging. As an application example, various types and sizes of MPs are identified through Raman imaging combined with a machine learning algorithm. The system has excellent performance with a spatial resolution of 2 µm and spectral resolution of 4 cm-1. Compared to traditional point-scanning Raman imaging systems, the detection speed is improved by 2 orders of magnitude. The pervasive nature of MPs results in their infiltration into the food chain, raising concerns for human health due to the potential for chemical leaching and the introduction of persistent organic pollutants. We conducted a series of experiments on various types and sizes of MPs. The system can give a classification accuracy of 98% for seven different types of plastics, and Raman imaging and species identification for MPs as small as 1 µm in diameter were achieved. We also identified toxic and harmful substances remaining in plastics, such as Dioctyl Phthalate (DOP) residues. This demonstrates a strong performance in microplastic species identification, size recognition and identification of hazardous substance contamination in microplastics.

Co-registered optical coherence tomography and X-ray angiography for the prediction of fractional flow reserve

Cardiovascular disease (CVD) stands as the leading global cause of mortality, and coronary artery disease (CAD) has the highest prevalence, contributing to 42% of these fatalities. Recognizing the constraints inherent in the anatomical assessment of CAD, Fractional Flow Reserve (FFR) has emerged as a pivotal functional diagnostic metric. Herein, we assess the potential of employing an ensemble approach with deep neural networks (DNN) to predict invasively measured Fractional Flow Reserve (FFR) using raw anatomical data extracted from both optical coherence tomography (OCT) and X-ray coronary angiography (XA). In this study, we used a challenging dataset, with 46% of the lesions falling within the FFR range of 0.75 to 0.85. Despite this complexity, our model achieved an accuracy of 84.3%, demonstrating a sensitivity of 87.5% and a specificity of 81.4%. Our results demonstrate that incorporating both OCT and XA signals, co-registered, as inputs for the DNN model leads to an important increase in overall accuracy.

Enhanced 3-D asynchronous correlation data preprocessing method for Raman spectroscopy of Chinese handmade paper

We have developed a novel 3D asynchronous correlation method (3D-ACM) designed for the classification and identification of Chinese handmade paper samples using Raman spectra and machine learning. The 3D-ACM approach involves two rounds of tensor product and Hilbert transform operations. In the tensor product process, the outer product of the spectral data from different samples within the same category is computed, establishing inner connections among all samples within that category. The Hilbert transform introduces a 90-degree phase shift, resulting in a true three-dimensional spectral data structure. This expansion significantly increases the number of equivalent frequency points and samples within each category. This enhancement substantially boosts spectral resolution and reveals more hidden information within the spectral data. To maximize the potential of 3D-ACM, we employed six machine learning models: principal component analysis (PCA) with linear regression (LR), support vector machine (SVM) with LR, k-Nearest Neighbors (KNN), random forest (RF), and convolutional neural network (CNN). When applied to the 3D-ACM data preprocessing method, R-squared values of PLS-LR, KNN, RF and CNN supervised models, approached or equaled 1. This indicates exceptional performance comparable to unsupervised models like PCA. 3D-ACM stands as a versatile mathematical technique not confined to spectral data. It also eliminates the necessity for additional experimental setups or external control conditions, distinct from traditional two-dimensional correlation spectroscopy. Moreover, it preserves the original experimental data, setting it apart from conventional data preprocessing methods. This positions 3D-ACM as a promising tool for future material classification and identification in conjunction with machine learning.

Application of Data Fusion in Traditional Chinese Medicine: A Review

Traditional Chinese medicine is characterized by numerous chemical constituents, complex components, and unpredictable interactions among constituents. Therefore, a single analytical technique is usually unable to obtain comprehensive chemical information. Data fusion is an information processing technology that can improve the accuracy of test results by fusing data from multiple devices, which has a broad application prospect by utilizing chemometrics methods, adopting low-level, mid-level, and high-level data fusion techniques, and establishing final classification or prediction models. This paper summarizes the current status of the application of data fusion strategies based on spectroscopy, mass spectrometry, chromatography, and sensor technologies in traditional Chinese medicine (TCM) in light of the latest research progress of data fusion technology at home and abroad. It also gives an outlook on the development of data fusion technology in TCM analysis to provide references for the research and development of TCM.

Nondestructive detection of Pleurotus geesteranus strain degradation based on micro-hyperspectral imaging and machine learning

In the production of edible fungi, the use of degraded strains in cultivation incurs significant economic losses. Based on micro-hyperspectral imaging and machine learning, this study proposes an early, nondestructive method for detecting different degradation degrees of Pleurotus geesteranus strains. In this study, an undegraded strain and three different degradation-level strains were used. During the mycelium growth, 600 micro-hyperspectral images were obtained. Based on the average transmittance spectra of the region of interest (ROI) in the range of 400-1000 nm and images at feature bands, feature spectra and images were extracted using the successive projections algorithm (SPA) and the deep residual network (ResNet50), respectively. Different feature input combinations were utilized to establish support vector machine (SVM) classification models. Based on the results, the spectra-input-based model performed better than the image-input-based model, and feature extraction improved the classification results for both models. The feature-fusion-based SPA+ResNet50-SVM model was the best; the accuracy rate of the test set was up to 90.8%, which was better than the accuracy rates of SPA-SVM (83.3%) and ResNet50-SVM (80.8%). This study proposes a nondestructive method to detect the degradation of Pleurotus geesteranus strains, which could further inspire new methods for the phenotypic identification of edible fungi.

Tensor product based 2-D correlation data preprocessing methods for Raman spectroscopy of Chinese handmade paper

The paper introduces two new methods, namely the cross correlation method (CCM) and two-dimensional correlation method (TDCM), for preprocessing Raman spectroscopy data for analyzing Chinese handmade paper samples. CCM expands the spectral dimension from 1×N to 1×2N-1 by taking cross-correlation between two spectral data of the same category. TDCM includes two-dimensional synchronous correlation method (TDSCM) and two-dimensional asynchronous correlation method (TDACM), which expand the spectral dimension from 1×N to N×N by taking tensor products between two spectral data and between one spectral data and the Hilbert transformation of the other spectral data of the same category, respectively. The experimental data were preprocessed using baseline removal, CCM, TDSCM, and TDACM methods. Four machine learning models were employed to evaluate the effects of these methods: principal component analysis (PCA) combined with linear regression (LR), support vector machine (SVM) combined with LR, k-Nearest Neighbors (KNN), and random forest (RF). The results show that the R-squared values for the PCA model were nearly 1 for all types of data, indicating high accuracy. However, for SVM-LR, KNN, and RF models, the R-squared values were sorted in the order of raw data, baseline removal data, CCM, TDSCM, and TDACM preprocessed data. The R-squared values of KNN and RF machine learning models for TDACM preprocessed data were approaching 1, indicating that the accuracy of machine learning was significantly improved by nearly 100%. This has led to a remarkable improvement in the accuracy of supervised models such as KNN and RF, bringing them closer to the level of unsupervised models such as PCA.

Quality assessment of traditional Chinese medicine based on data fusion combined with machine learning: A review

The authenticity and quality of traditional Chinese medicine (TCM) directly impact clinical efficacy and safety. Quality assessment of traditional Chinese medicine (QATCM) is a global concern due to increased demand and shortage of resources. Recently, modern analytical technologies have been extensively investigated and utilized to analyze the chemical composition of TCM. However, a single analytical technique has some limitations, and judging the quality of TCM only from the characteristics of the components is not enough to reflect the overall view of TCM. Thus, the development of multi-source information fusion technology and machine learning (ML) has further improved QATCM. Data information from different analytical instruments can better understand the connection between herbal samples from multiple aspects. This review focuses on the use of data fusion (DF) and ML in QATCM, including chromatography, spectroscopy, and other electronic sensors. The common data structures and DF strategies are introduced, followed by ML methods, including fast-growing deep learning. Finally, DF strategies combined with ML methods are discussed and illustrated for research on applications such as source identification, species identification, and content prediction in TCM. This review demonstrates the validity and accuracy of QATCM-based DF and ML strategies and provides a reference for developing and applying QATCM methods.

Detection of Rice Fungal Spores Based on Micro- Hyperspectral and Microfluidic Techniques

As rice is one of the world's most important food crops, protecting it from fungal diseases is very important for agricultural production. At present, it is difficult to diagnose rice fungal diseases at an early stage using relevant technologies, and there are a lack of rapid detection methods. This study proposes a microfluidic chip-based method combined with microscopic hyperspectral detection of rice fungal disease spores. First, a microfluidic chip with a dual inlet and three-stage structure was designed to separate and enrich Magnaporthe grisea spores and Ustilaginoidea virens spores in air. Then, the microscopic hyperspectral instrument was used to collect the hyperspectral data of the fungal disease spores in the enrichment area, and the competitive adaptive reweighting algorithm (CARS) was used to screen the characteristic bands of the spectral data collected from the spores of the two fungal diseases. Finally, the support vector machine (SVM) and convolutional neural network (CNN) were used to build the full-band classification model and the CARS filtered characteristic wavelength classification model, respectively. The results showed that the actual enrichment efficiency of the microfluidic chip designed in this study on Magnaporthe grisea spores and Ustilaginoidea virens spores was 82.67% and 80.70%, respectively. In the established model, the CARS-CNN classification model is the best for the classification of Magnaporthe grisea spores and Ustilaginoidea virens spores, and its F1-core index can reach 0.960 and 0.949, respectively. This study can effectively isolate and enrich Magnaporthe grisea spores and Ustilaginoidea virens spores, providing new methods and ideas for early detection of rice fungal disease spores.

Determination of geographic origins and types of Lindera aggregata samples using a portable short-wave infrared hyperspectral imager

A portable short-wavelength infrared microscope hyperspectral imager (SMHI) combined with machine learning algorithms for the purpose of classifying geographical origins as well as root types of Lindera aggregata is developed. The spectral range of the SMHI system is 1090-1820 nm (5500-9100 cm-1) with spectral and spatial resolutions of 4 nm and 27.3 μm, respectively. Utilizing PCA-RF algorithms, the geographic origin of tuberous roots and leaves from five different origins were classified with accuracies of 97.5% and 97.8%, respectively. In addition, spatial identification of tuberous root and taproot tubers in a mixed sample was done with an accuracy of 98.98%. The accuracy of origin classification and spatial identification are high enough which indicate the significant potential of applying SMHI system into the non-invasive spatial mapping and rapid quality assessment of medicinal herbs.

The complete chloroplast genomes of Tetrastigma hemsleyanum (Vitaceae) from different regions of China: molecular structure, comparative analysis and development of DNA barcodes for its geographical origin discrimination

BACKGROUND

Tetrastigma hemsleyanum is a valuable traditional Chinese medicinal plant widely distributed in the subtropical areas of China. It belongs to the Cayratieae tribe, family Vitaceae, and exhibited significant anti-tumor and anti-inflammatory activities. However, obvious differences were observed on the quality of T. hemsleyanum root from different regions, requiring the discrimination strategy for the geographical origins.

RESULT

This study characterized five complete chloroplast (cp) genomes of T. hemsleynum samples from different regions, and conducted a comparative analysis with other representing species from family Vitaceae to reveal the structural variations, informative markers and phylogenetic relationships. The sequenced cp genomes of T. hemsleyanum exhibited a conserved quadripartite structure with full length ranging from 160,124 bp of Jiangxi Province to 160,618 bp of Zhejiang Province. We identified 112 unique genes (80 protein-coding, 28 tRNA and 4 rRNA genes) in the cp genomes of T. hemsleyanum with highly similar gene order, content and structure. The IR contraction/expansion events occurred on the junctions of ycf1, rps19 and rpl2 genes with different degrees, causing the differences of genome sizes in T. hemsleyanum and Vitaceae plants. The number of SSR markers discovered in T. hemsleyanum was 56-57, exhibiting multiple differences among the five geographic groups. Phylogenetic analysis based on conserved cp genome proteins strongly grouped the five T. hemsleyanum species into one clade, showing a sister relationship with T. planicaule. Comparative analysis of the cp genomes from T. hemsleyanum and Vitaceae revealed five highly variable spacers, including 4 intergenic regions and one protein-coding gene (ycf1). Furthermore, five mutational hotspots were observed among T. hemsleyanum cp genomes from different regions, providing data for designing DNA barcodes trnL and trnN. The combination of molecular markers of trnL and trnN clustered the T. hemsleyanum samples from different regions into four groups, thus successfully separating specimens of Sichuan and Zhejiang from other areas.

CONCLUSION

Our study obtained the chloroplast genomes of T. hemsleyanum from different regions, and provided a potential molecular tracing tool for determining the geographical origins of T. hemsleyanum, as well as important insights into the molecular identification approach and and phylogeny in Tetrastigma genus and Vitaceae family.

Towards a Deep-Learning Approach for Prediction of Fractional Flow Reserve from Optical Coherence Tomography

Cardiovascular disease (CVD) is the number one cause of death worldwide, and coronary artery disease (CAD) is the most prevalent CVD, accounting for 42% of these deaths. In view of the limitations of the anatomical evaluation of CAD, Fractional Flow Reserve (FFR) has been introduced as a functional diagnostic index. Herein, we evaluate the feasibility of using deep neural networks (DNN) in an ensemble approach to predict the invasively measured FFR from raw anatomical information that is extracted from optical coherence tomography (OCT). We evaluate the performance of various DNN architectures under different formulations: regression, classification—standard, and few-shot learning (FSL) on a dataset containing 102 intermediate lesions from 80 patients. The FSL approach that is based on a convolutional neural network leads to slightly better results compared to the standard classification: the per-lesion accuracy, sensitivity, and specificity were 77.5%, 72.9%, and 81.5%, respectively. However, since the 95% confidence intervals overlap, the differences are statistically not significant. The main findings of this study can be summarized as follows: (1) Deep-learning (DL)-based FFR prediction from reduced-order raw anatomical data is feasible in intermediate coronary artery lesions; (2) DL-based FFR prediction provides superior diagnostic performance compared to baseline approaches that are based on minimal lumen diameter and percentage diameter stenosis; and (3) the FFR prediction performance increases quasi-linearly with the dataset size, indicating that a larger train dataset will likely lead to superior diagnostic performance.

4D dual-mode staring hyperspectral-depth imager for simultaneous spectral sensing and surface shape measurement

A 4D dual-mode staring hyperspectral-depth imager (DSHI), which acquire reflectance spectra, fluorescence spectra, and 3D structural information by combining a staring hyperspectral scanner and a binocular line laser stereo vision system, is introduced. A 405 nm laser line generated by a focal laser line generation module is used for both fluorescence excitation and binocular stereo matching of the irradiated line region. Under the configuration, the two kinds of hyperspectral data collected by the hyperspectral scanner can be merged into the corresponding points in the 3D model, forming a dual-mode 4D model. The DSHI shows excellent performance with spectral resolution of 3 nm, depth accuracy of 26.2 µm. Sample experiments on a fluorescent figurine, real and plastic sunflowers and a clam are presented to demonstrate system's with potential within a broad range of applications such as, e.g., digital documentation, plant phenotyping, and biological analysis.

Wavelength selection of multispectral imaging for oil palm fresh fruit ripeness classification

Multispectral imaging has been recently proposed for high-speed sorting and grading machine vision of fruits. It is a prospective method applied in yet traditional sorting and grading of oil palm fresh fruit bunches (FFB). The ripeness of oil palm FFBs determines the quality of crude palm oil (CPO). Implementation of multispectral imaging for the task needs wavelength selection from hyperspectral datasets. This study aimed to obtain the optimum wavelengths and use them for oil palm FFB classification based on three ripeness levels. We have selected eight optimum wavelengths using principal component analysis (PCA) regression which represented the ripeness levels.

Automated defect detection and classification for fiber-optic coil based on wavelet transform and self-adaptive GA-SVM

The quality monitoring of fiber-optic coil (FOC) in winding systems is usually done manually. Aiming at the problem of inefficient and low accuracy of manual detection, this article is dedicated to researching a defect detection framework based on machine vision, which provides a reliable method for automatic defect detection of FOC. For this purpose, a defect detection scheme that integrates wavelet transform and nonlocal means filtering is proposed to accurately locate the defect region. Then, based on the features constructed by wavelet coefficients, a support vector machine (SVM) is used as the classifier. Additionally, a self-adaptive genetic algorithm is proposed to optimize the parameters of the SVM to form the final classifier. Through experiments on the data set obtained by our designed imaging system, the results show that our method has good defect detection performance and high classification accuracy, which provides an optimal solution for the automatic detection of FOC.

Fluorescence hyperspectral imaging system for analysis and visualization of oil sample composition and thickness

In this paper, a compact fluorescence hyperspectral imaging system based on a prism-grating-prism (PGP) structure is designed. Its spectrometer spectral range is 400-1000 nm with a spectral resolution of 2.5 nm, and its weight is less than 1.7 kg. The PGP imaging spectrometer combines the technical advantages of prism and grating, by not only using six lenses for imaging and collimation to realize the dual telecentres of object and image but also having a "straight cylinder" structure, which makes the installation and adjustment simple, compact, and stable. By the push-broom method, we obtained the three-dimensional cubic data of different oil products. By normalization processing, minimum noise separation transformation processing, visualization processing, and support vector machine classification processing of different oil fluorescence hyperspectral data, we demonstrate that the fluorescence hyperspectral imaging system can identify different kinds of oil and recognize the oil film thickness. The fluorescence hyperspectral imaging system can be used in oil spill detection, resource exploration, natural disaster monitoring, environmental pollution assessment, and many other fields.

WisdomModel: convert data into wisdom

Purpose

Traditional classification algorithms always have an incorrect prediction. As the misclassification rate increases, the usefulness of the learning model decreases. This paper presents the development of a wisdom framework that reduces the error rate to less than 3% without human intervention.

Design/methodology/approach

The proposed WisdomModel consists of four stages: build a classifier, isolate the misclassified instances, construct an automated knowledge base for the misclassified instances and rectify incorrect prediction. This approach will identify misclassified instances by comparing them against the knowledge base. If an instance is close to a rule in the knowledge base by a certain threshold, then this instance is considered misclassified.

Findings

The authors have evaluated the WisdomModel using different measures such as accuracy, recall, precision, f-measure, receiver operating characteristics (ROC) curve, area under the curve (AUC) and error rate with various data sets to prove its ability to generalize without human involvement. The results of the proposed model minimize the number of misclassified instances by at least 70% and increase the accuracy of the model minimally by 7%.

Originality/value

This research focuses on defining wisdom in practical applications. Despite of the development in information system, there is still no framework or algorithm that can be used to extract wisdom from data. This research will build a general wisdom framework that can be used in any domain to reach wisdom.