Development of a classification model for Cynanchum wilfordii and Cynanchum auriculatum using convolutional neural network and local interpretable model-agnostic explanation technology

Cynanchum wilfordii is a perennial tuberous root in the Asclepiadaceae family that has long been used medicinally. Although C. wilfordii is distinct in origin and content from Cynancum auriculatum, a genus of the same species, it is difficult for the public to recognize because the ripe fruit and root are remarkably similar. In this study, images were collected to categorize C. wilfordii and C. auriculatum, which were then processed and input into a deep-learning classification model to corroborate the results. By obtaining 200 photographs of each of the two cross sections of each medicinal material, approximately 800 images were employed, and approximately 3200 images were used to construct a deep-learning classification model via image augmentation. For the classification, the structures of Inception-ResNet and VGGnet-19 among convolutional neural network (CNN) models were used, with Inception-ResNet outperforming VGGnet-19 in terms of performance and learning speed. The validation set confirmed a strong classification performance of approximately 0.862. Furthermore, explanatory properties were added to the deep-learning model using local interpretable model-agnostic explanation (LIME), and the suitability of the LIME domain was assessed using cross-validation in both situations. Thus, artificial intelligence may be used as an auxiliary metric in the sensory evaluation of medicinal materials in future, owing to its explanatory ability.

Terahertz Spectroscopy and Imaging Techniques for Herbal Medicinal Plants Detection: A Comprehensive Review

Herbal medicine (HM), derived from various therapeutic plants, has garnered considerable attention for its remarkable effectiveness in treating diseases. However, numerous issues including improved varieties selection, hazardous residue detection, and concoction management affect herb quality throughout the manufacturing process. Therefore, a practical, rapid, nondestructive detection technology is necessary. Terahertz (THz) spectroscopy, with low energy, penetration, and fingerprint features, becomes preferable method for herb quality appraisal. There are three parts in our review. THz techniques, data processing, and modeling methods were introduced in Part I. Three primary applications (authenticity, composition and active ingredients, and origin detection) of THz in medicinal plants quality detection in industrial processing and marketing were detailed in Part II. A thorough investigation and outlook on the well-known applications and advancements of this field were presented in Part III. This review aims to bring new enlightenment to the in-depth THz application research in herbal medicinal plants.

Cynanchum auriculatum Royle ex Wight., Cynanchum bungei Decne. and Cynanchum wilfordii (Maxim.) Hemsl.: Current Research and Prospects

Cynanchum auriculatum Royle ex Wight. (CA), Cynanchum bungei Decne. (CB) and Cynanchum wilfordii (Maxim.) Hemsl. (CW) are three close species belonging to the Asclepiadaceous family, and their dry roots as the bioactive part have been revealed to exhibit anti-tumor, neuroprotection, organ protection, reducing liver lipid and blood lipid, immunomodulatory, anti-inflammatory, and other activities. Until 2021, phytochemistry investigations have uncovered 232 compounds isolated from three species, which could be classified into C21-steroids, acetophenones, terpenoids, and alkaloids. In this review, the morphology characteristics, species identification, and the relationship of botany, extraction, and the separation of chemical constituents, along with the molecular mechanism and pharmacokinetics of bioactive constituents of three species, are summarized for the first time, and their phytochemistry, pharmacology, and clinical safety are also updated. Moreover, the direction and limitation of current research on three species is also discussed.

Polygonum multiflorum: Recent updates on newly isolated compounds, potential hepatotoxic compounds and their mechanisms

ETHNOPHARMACOLOGICAL RELEVANCE

Polygonum multiflorum Thunb.(PM), (known as Heshouwu () in China) is one of the most important and well mentioned Chinese medicinal herbs in the literature for its use in blackening hair, nourishing liver and kidney, anti-aging, anti-hyperlipidemia, antioxidant, anti-inflammatory, anticancer, hepatoprotection, cardio-protection and improving age-related cognitive dysfunction. The purpose of this review is to give a comprehensive and recent update on PM: new compounds or isolated for the first time, potential hepatotoxic compounds and their mechanisms. Moreover, future perspectives and challenges in the future study of this plant are conversed which will make a new base for further study on PM.

MATERIALS AND METHODS

A comprehensive review of relevant published literature on PM using the scientific databases SCOPUS, PubMed, and Science Direct was done.

RESULTS

PM is broadly produced in many provinces of China and well known in other Eastern Asian Countries for its ethno-medical uses. Previous phytochemical investigation of PM had led to the isolation of more than 175 compounds including recently isolated 70 new compounds. Most of the new compounds isolated after 2015 are majorly dianthrone glycosides and stilbene glycosides. Processing has also a significant effect on chemical composition, pharmacological activities, and toxicity of PM. PM-induced liver injury is increasing after the first report in Hong Kong in 1996. Hepatotoxicity of PM was constantly reported in Japan, Korea, China, Australia, Britain, Italy, and other countries although its toxicity is related to idiosyncratic hepatotoxicity. More interestingly, although there is indispensable interest to predict idiosyncratic hepatotoxicity of PM and understand its mechanisms, the responsible hepatotoxic compounds and mechanisms of liver damage induced by PM are still not clear. There is a big controversy on the identification of the most responsible constituent. Anthraquinone and stilbene compounds in PM, mainly emodine and TSG are mentioned in the literature to be the main responsible hepatotoxic compounds. However, comparing the two compounds, which one is the more critical toxic agent for PM-induced hepatotoxicity is not well answered. Affecting different physiological and metabolic pathways such as oxidative phosphorylation and TCA cycle pathway, metabolic pathways, bile acid excretion pathway and genetic polymorphisms are among the mechanisms of hepatotoxicity of PM.

CONCLUSION

Deeper and effective high throughput experimental studies are still research hotspots to know the most responsible constituent and the mechanism of PM-induced hepatotoxicity.

Preliminary Quality Evaluation and Characterization of Phenolic Constituents in Cynanchi Wilfordii Radix

A new phenolic compound, 2-O-β-laminaribiosyl-4-hydroxyacetophenone (1), was isolated from Cynanchi Wilfordii Radix (CWR, the root of Cynanchum wilfordii Hemsley), along with 10 known aromatic compounds, including cynandione A (2), bungeisides-C (7) and –D (8), p-hydroxyacetophenone (9), 2′,5′-dihydroxyacetophenone (10), and 2′,4′-dihydroxyacetophenone (11). The structure of the new compound (1) was elucidated using spectroscopic methods and chemical methods. The structure of cynandione A (2), including a linkage mode of the biphenyl parts that remained uncertain, was unambiguously confirmed using the 2D ¹³C–¹³C incredible natural abundance double quantum transfer experiment (INADEQUATE) spectrum. Additionally, health issues related to the use of Cynanchi Auriculati Radix (CAR, the root of Cynanchum auriculatum Royle ex Wight) instead of CWR have emerged. Therefore, constituents present in methanolic extracts of commercially available CWRs and CARs were examined using UV-sensitive high-performance liquid chromatography (HPLC), resulting in common detection of three major peaks ascribed to cynandione A (2), p-hydroxyacetophenone (9), and 2′,4′-dihydroxyacetophenone (11). Thus, to distinguish between these ingredients, a thin-layer chromatography (TLC) method, combined with only UV irradiation detection, focusing on wilfosides C1N (12) and K1N (13) as marker compounds characteristic of CAR, was performed. Furthermore, we propose this method as a simple and convenient strategy for the preliminary distinction of CWR and CAR to ensure the quality and safety of their crude drugs.