Raman spectroscopy for on-line monitoring of botanical extraction process using convolutional neural network with background subtraction.

Aqueous extraction is the most common and cost-effective means of obtaining active ingredients from medicinal plants. However, botanical extracts generally contain high pigment content and complex chemical composition posing a challenge for the process analysis of aqueous extraction. Here, we employed Raman spectroscopy to monitor the physical and chemical properties during the extraction process using convolution neural network (CNN) with background subtraction. Real-time spectra were first preprocessed to eliminate fluorescence background interference. Next, two types of CNN models, the one-dimensional CNN (1D-CNN) based on one preprocessing method, and two-dimensional CNN (2D-CNN) based on a concatenation of differentially pretreated data blocks, were used to receive the preprocessed spectra data. Two case studies were conducted for 1D- and 2D-CNN: the extraction of Aurantii fructus, and the co-extraction of Radix Salvia miltiorrhiza and Rhizoma Ligusticum chuanxiong. Furthermore, partial least squares (PLS) models and sequential preprocessing through orthogonalization (SPORT) models were developed and compared with 1D-CNN and 2D-CNN, respectively. CNN-based methods were superior to other models in terms of prediction accuracy, with 2D-CNN yielding the best results. These results indicated that preprocessing and CNN methods were highly complementary, and could effectively remove the fluorescence effect and artefacts introduced by pretreatment in spectral profile. To the best of our knowledge, this is the first study to demonstrate that a combination of preprocessing and CNN leads to improved prediction performance of analytes when using Raman spectroscopy for online monitoring high-pigmented samples.

An online, non-destructive method for simultaneously detecting chemical, biological, and physical properties of herbal injections using hyperspectral imaging with artificial intelligence.

Botanical drugs hold great potential to prevent and treat complex diseases. Quality control is essential in ensuring the safety, efficacy, and therapeutic consistency of these drug products. The quality of a botanical drug product can be assessed using a variety of analytical methods based on criteria that judge the identity, strength, purity, and potency. However, most of these methods are developed on separate analytical platforms, and few approaches are available for in-process monitoring of multiple quality properties in a non-destructive manner. Here, we present a hyperspectral imaging-based strategy for online measurement of physical, chemical, and biological properties of botanical drugs using artificial intelligence algorithms. An end-to-end convolutional neural network (CNN) model was established to accurately determine phytochemicals and bioactivities based on the spectra. Besides, a new dual-scale anomaly (DSA) detection algorithm was proposed for visible particle inspection based on the images. The strategy was exemplified on Shuxuening Injection, a Ginkgo biloba-derived drug used in the treatment of cerebrovascular and cardiovascular diseases. Four quality metrics of the injection, including total flavonol, total ginkgolides, antioxidant activity, and anticoagulant activity, were successfully predicted by the CNN model with validation R2 of 0.922, 0.921, 0.880, and 0.913 respectively, showing better performance than the other models. Unqualified samples with visible particles could be detected by DSA with a low false alarm rate of 9.38 %. Chromaticity results indicated that the inter-company variations of color were significant, while intra-company variations were relatively small. This demonstrates a real application of integrating hyperspectral imaging with artificial intelligence to provide a rapid, accurate, and non-destructive approach for process analysis of botanical drugs.

[Identification of Armeniacae Semen Amarum and Persicae Semen from different origins based on near infrared hyperspectral imaging technology].

In order to establish a rapid and non-destructive evaluation method for the identification of Armeniacae Semen Amarum and Persicae Semen from different origins, the spectral information of Armeniacae Semen Amarum and Persicae Semen in the range of 898-1 751 nm was collected based on hyperspectral imaging technology. Armeniacae Semen Amarum and Persicae Semen from different origins were collected as research objects, and a total of 720 Armeniacae Semen Amarum samples and 600 Persicae Semen samples were used for authenticity discrimination. The region of interest(ROI) and the average reflection spectrum in the ROI were obtained, followed by comparing five pre-processing methods. Then, partial least squares discriminant analysis(PLS-DA), support vector machine(SVM), and random forest(RF) method were established for classification models, which were evaluated by the confusion matrix of prediction results and receiver operating characteristic curve(ROC). The results showed that in the three sample sets, the se-cond derivative pre-processing method and PLS-DA were the best model combinations. The classification accuracy of the test set under the 5-fold cross-va-lidation was 93.27%, 96.19%, and 100.0%, respectively. It was consistent with the confusion matrix of the predicted results. The area under the ROC curve obtained the highest values of 0.992 3, 0.999 6, and 1.000, respectively. The study revealed that the near-infrared hyperspectral imaging technology could accurately identify the medicinal materials of Armeniacae Semen Amarum and Persicae Semen from different origins and distinguish the authentication of these two varieties.

[Fusion of spectrum and image features to identify Glycyrrhizae Radix et Rhizoma from different origins based on hyperspectral imaging technology].

To identify Glycyrrhizae Radix et Rhizoma from different geographical origins, spectrum and image features were extracted from visible and near-infrared(VNIR, 435-1 042 nm) and short-wave infrared(SWIR, 898-1 751 nm) ranges based on hyperspectral imaging technology. The spectral features of Glycyrrhizae Radix et Rhizoma samples were extracted from hyperspectral data and denoised by a variety of pre-processing methods. The classification models were established by using Partial Least Squares Discriminate Analysis(PLS-DA), Support Vector Classification(SVC) and Random Forest(RF). Meanwhile, Gray-Level Co-occurrence matrix(GLCM) was employed to extract textural variables. The spectrum and image data were implemented from three dimensions, including VNIR and SWIR fusion, spectrum and image fusion, and comprehensive data fusion. The results indicated that the spectrum in SWIR range performed better classification accuracy than VNIR range. Compared with other four pre-processing methods, the second derivative method based on Savitzky-Golay(SG) smoothing exhibited the best performance, and the classification accuracy of PLS-DA and SVC models were 93.40% and 94.11%, separately. In addition, the PLS-DA model was superior to SVC and RF models in terms of classification accuracy and model generalization capability, which were evaluated by confusion matrix and receiver operating characteristic curve(ROC). Comprehensive data fusion on SPA bands achieved a classification accuracy of 94.82% with only 28 bands. As a result, this approach not only greatly improved the classification efficiency but also maintained its accuracy. The hyperspectral imaging system, a non-invasively, intuitively and quickly identify technology, could effectively distinguish Glycyrrhizae Radix et Rhizoma samples from different origins.

A Hyperspectral Imaging Approach for Classifying Geographical Origins of Rhizoma Atractylodis Macrocephalae Using the Fusion of Spectrum-Image in VNIR and SWIR Ranges (VNIR-SWIR-FuSI).

Hyperspectral data processing technique has gained increasing interests in the field of chemical and biomedical analysis. However, appropriate approaches to fusing features of hyperspectral data-cube are still lacking. In this paper, a new data fusion approach was proposed and applied to discriminate Rhizoma Atractylodis Macrocephalae (RAM) slices from different geographical origins using hyperspectral imaging. Spectral and image features were extracted from hyperspectral data in visible and near-infrared (VNIR, 435-1042 nm) and short-wave infrared (SWIR, 898-1751 nm) ranges, respectively. Effective wavelengths were extracted from pre-processed spectral data by successive projection algorithm (SPA). Meanwhile, gray-level co-occurrence matrix (GLCM) and gray-level run-length matrix (GLRLM) were employed to extract textural variables. The fusion of spectrum-image in VNIR and SWIR ranges (VNIR-SWIR-FuSI) was implemented to integrate those features on three fusion dimensions, i.e., VNIR and SWIR fusion, spectrum and image fusion, and all data fusion. Based on data fusion, partial least squares-discriminant analysis (PLS-DA) and support vector machine (SVM) were utilized to establish calibration models. The results demonstrated that VNIR-SWIR-FuSI could achieve the best accuracies on both full bands (97.3%) and SPA bands (93.2%). In particular, VNIR-SWIR-FuSI on SPA bands achieved a classification accuracy of 93.2% with only 23 bands, which was significantly better than those based on spectra (80.9%) or images (79.7%). Thus it is more rapid and possible for industry applications. The current study demonstrated that hyperspectral imaging technique with data fusion holds the potential for rapid and nondestructive sorting of traditional Chinese medicines (TCMs).

[Studies on the methodology for quality control in Chinese medicine manufacturing process based on knowledge graph].

According to the requirements for developing the quality control technology in Chinese medicine( CM) manufacturing process and the practical scenarios in applying a new generation of artificial intelligence to CM industry,we present a method of constructing the knowledge graph( KG) for CM manufacture to solve key problems about quality control in CM manufacturing process.Based on the above,a " pharmaceutical industry brain" model for CM manufacture has been established. Further,we propose founding the KG-based methodology for quality control in CM manufacturing process,and briefly describe the design method,system architecture and main functions of the KG system. In this work,the KG for manufacturing Shuxuening Injection( SXNI) was developed as a demonstration study. The KG version 1. 0 platform for intelligent manufacturing SXNI has been built,which could realize technology leap of the quality control system in CM manufacturing process from perceptual intelligence to cognitive intelligence.