In recent years, with the increasing societal focus on drug quality and safety, quality issues have become a major challenge faced by the pharmaceutical industry, directly impacting consumer health and market trust. By combining multispectral imaging technology with machine learning, it is possible to achieve rapid, non-destructive, and precise detection of traditional Chinese medicine(TCM) preparations, thereby revolutionizing traditional detection methods and developing more convenient and automated solutions. This paper provides a comprehensive review of the current applications of rapid, non-destructive detection techniques based on machine learning algorithms in the field of TCM preparations. It analyzed the principles and advantages of commonly used rapid, non-destructive detection techniques, offering a reference for the application and promotion of these technologies in TCM preparation detection. Additionally, this paper explored various data preprocessing techniques, operational processes, and machine learning algorithms to enhance data utilization efficiency. Finally, it focused on the challenges of applying machine learning in TCM preparation detection and offered corresponding recommendations, providing guidance for the future integration of machine learning with rapid, non-destructive detection techniques in practical production.
Machine Learning ; Drugs, Chinese Herbal/analysis* ; Medicine, Chinese Traditional/methods* ; Humans ; Quality Control

The construction method and simulation parameter settings for the discrete element model of Jianwei Xiaoshi Granules, as the primary material of Jianwei Xiaoshi Tablets, are not yet clear. The accuracy of the simulation model significantly influences the dynamic response characteristics between granules. Therefore, it is necessary to calibrate the parameters to improve the accuracy of the simulation parameters. Using the repose angle of Jianwei Xiaoshi Granules as the response value, the response surface methodology was employed to optimize and calibrate the discrete element parameters. Physical experiments were conducted to determine the physical properties of Jianwei Xiaoshi Granules. Based on the Hertz-Mindlin with Johnson-Kendall-Roberts(JKR) V2 model and virtual simulation methods, a repose angle determination model was constructed in EDEM software. The repose angle was measured using image analysis and numerical fitting methods. The Plackett-Burman experiment was used to screen the initial parameters for significance in the discrete element simulation. The significant parameters were then subjected to a steepest ascent experiment to determine the optimal parameter range. Furthermore, based on the Box-Behnken experiment, a second-order regression equation between significant parameters and repose angle was established, with the repose angle of 37.64° in the physical experiment as the target value. The regression equation was optimized and solved. The significance screening experiment revealed that the granule-granule static friction coefficient, granule-granule rolling friction, and granule-steel plate rolling friction of Jianwei Xiaoshi Granules significantly influenced the simulated repose angle. The optimal parameter combination was found to be 0.330, 0.222, and 0.229. The simulation results with this optimal parameter combination showed that there was no significant difference between the simulated repose angle and the repose angle obtained in the physical experiment, with a relative error of 0.05%, which further validated the reliability of the calibrated discrete element parameters for Jianwei Xiaoshi Granules.
Drugs, Chinese Herbal/chemistry* ; Calibration ; Computer Simulation

Identification of critical material attributes(CMAs) is a key issue in the quality control of large-scale TCM products like Jianwei Xiaoshi Tablets. This study focuses on the granules of Jianwei Xiaoshi Tablets, using tablet tensile strength as the primary quality attribute. A method for identifying the CMAs and a design space for the granules were established, along with a predictive model for the granule CMAs based on Fourier transform near-infrared spectroscopy(FT-NIR). First, granules of Jianwei Xiaoshi Tablets with different properties were prepared using a partial factorial design method from the design of experiments(DOE). The powder properties of the granules were measured. An orthogonal partial least squares(OPLS) model was established to correlate the powder properties with tensile strength. Based on the characteristics of the comprehensive variables extracted by OPLS, the independent variables with the greatest explanatory power for tensile strength were identified. FT-NIR technology was then employed to establish a predictive model for the granule CMAs. The final CMAs identified were hygroscopicity, moisture content, D_(50), collapse angle, mass flow rate, and tapped density. The coefficients of determination of the prediction set(R■) and relative percentage deviation(RPD) of the prediction set for flowability, D_(50), and moisture content were 0.891, 0.994, and 0.998; and 2.97, 12.4, and 20.7, respectively. The established OPLS model clearly identified the impact of various factors on tensile strength, demonstrating good fit results. The model exhibited high prediction accuracy and can be used for the rapid and accurate determination of CMAs in granules of Jianwei Xiaoshi Tablets.
Drugs, Chinese Herbal/chemistry* ; Tablets/chemistry* ; Tensile Strength ; Quality Control ; Spectroscopy, Fourier Transform Infrared ; Spectroscopy, Near-Infrared

Succus Bambusae， regarded as the holy medicine for the treatment of phlegm by traditional Chinese medicine experts， has the functions of clearing heat and resolving phlegm， resuscitating and relieving convulsions. In clinical practice， it is mainly used to treat phlegm-heat cough， phlegm yellow and shortness of breath， stroke with exuberant phlegm and other syndromes. The research of its processing procedure and quality standard is the key issue to realize the process modernization and quality standardization of this traditional medicine resource. On the basis of literature research， this paper conducted the herbal textual research on Succus Bambusae， sorted out its medicinal history， and compared the advantages and disadvantages of its traditional and modern preparation procedure. Based on the historical records， it is pointed out that there are still many problems in the traditional and modern processing procedure of Succus Bambusae， such as low-yield processing procedure and extensive production technology that cannot meet the requirements of large-scale green manufacturing. At present， there are a lot of literature on the comparative study of the origin， variety and processing procedure of Succus Bambusae， but most of them are limited to the chemical components， and lack of in-depth study on its pharmacodynamic substance basis and mechanism. Meanwhile， the pharmacological effects of Succus Bambusae are mostly limited to antitussive and expectorant， and few other pharmacological effects have been reported. Based on the above analysis， it is suggested that the follow-up research work can be carried out from these five aspects：①Clarifying the original source， enriching and perfecting the medicinal resources. ②Standardizing the processing procedure， establishing relevant quality standards. ③Clarifying the pharmacological mechanism， strengthening the basic research on the pharmacological effect and clinical application. ④Inheriting ancient processing， researching and developing special preparation equipment and standardizing production operation rules. ⑤Comprehensive utilization of the by-products in preparation process.

Objective:To explore the possible mechanism of Huangqintang in treating ulcerative colitis （UC）. Method:The animal model of UC was induced by dextran sodium sulfate （DSS）．The experimental animals were divided into control group， model group，Huangqintang low dose （4.55 g·kg^-1）， medium dose （9.1 g·kg^-1）， and high dose（18.2 g·kg^-1） groups. Intragastric administration was also given in the modeling process for 7 consecutive days. At the end of the 8th day， colon tissues were collected to measure colon length and mass， and calculate the colon mass index. Pathological changes were observed by hematoxylin-eosin （HE） staining. Serum iron content， superoxide dismutase （SOD）， glutathione （GSH）， catalase （CAT） and myeloperoxidase （MPO） were determined by biochemical assay. Western blot was used to detect the protein expression of glutathione peroxidase 4 （GSH-Px4）， long-chain acyl-CoA synthetase 4 （ACSL4） and ferritin heavy chain 1（FTH1）. The mRNA expression levels of tumor trotein 53 （P53） and solute carrier family 7 member 11 （SLC7A11） in colon tissues were detected by Real-time fluorescence quantitative polymerase chain reaction （Real-time PCR）. Result:The experimental studies showed that compared with normal group， serum MPO and iron content， ACSL4 protein level and relative P53 mRNA expression in the model group significantly increased （P<0.05）， while serum SOD， CAT， GSH content， GSH-Px4， FTH1 relative protein expression level and relative SLC7A11 mRNA expression in the model group significantly decreased （P<0.01）. Compared with model group， serum MPO and iron content， ACSL4 protein level and relative P53 mRNA expression significantly decreased （P<0.05）， while serum SOD， CAT， GSH content， GSH-Px4， FTH1 relative protein expression level and relative SLC7A11 mRNA expression significantly increased （P<0.05） after the intervention of Huangqintang， and the effect was most significant in the high-dose group （P<0.05）. The results of general condition， colon length， colon mass index and HE staining showed that Huangqintang could relieve clinical symptoms and histopathological changes in UC mice. Conclusion:These results indicated that Huangqintang had therapeutic effect on ulcerative colitis mice， and its mechanism might be related to inhibiting the oxidative stress and ferroptosis.

To explore the effect of Huangqin Decoction on ulcerative colitis(UC) pyroptosis, and to explain the mechanism of pyroptosis based on NOD-like receptor thermoprotein domain 3(NLRP3)/cysteine proteinase 1(caspase-1) pathway. The animal model of UC induced with 3% dextran sodium sulfate(DSS) was established. The experimental animals were divided into control group, model group, low-dose(4.55 g·kg~(-1)), medium-dose(9.1 g·kg~(-1)) and high-dose(18.2 g·kg~(-1)) Huangqin Decoction groups and salazosulfapyridine group(0.45 g·kg~(-1)). While modeling, intragastric administration was given for 7 consecutive days. On the 8 th day, the mice were euthanized, the colon length was collected, and the histopathological changes were observed by HE staining. The content of interleukin-18(IL-18) was observed by ELISA. The content of lactatedehydrogenase(LDH)was determined by microplate method. TUNEL assay kit was used to detect the cell death. The immunohistochemical staining was used to detect the expressions of NLRP3 and apoptosis-associated speck-like protein containing a CARD(ASC). Western blot was used to detect the expressions of interleukin-1β(IL-1β), caspase-1 and gasdermin D(GSDMD).The experimental study showed that compared with normal group, the LDH content, TUNEL positive staining, inflammatory factors(IL-18, IL-1β), and proteins associated with pyroptosis were significantly increased(P<0.05). Compared with model control group, the LDH content, TUNEL positive staining, inflammatory factors(IL-18, IL-1β), and proteins associated with pyroptosis were decreased, and these results were more significant in high-dose groups(P<0.05). The results of HE staining showed that Huangqin Decoction could improve the pathological changes of colon. Huangqin Decoction could inhibit UC cell pyroptosis, and the mechanism may be closely related to NLRP3/caspase-1 signaling pathway.
Animals ; Caspase 1/genetics* ; Colitis, Ulcerative/drug therapy* ; Mice ; NLR Family, Pyrin Domain-Containing 3 Protein/genetics* ; Pyroptosis ; Scutellaria baicalensis

OBJECTIVEExtraction of cepstral coefficients combined with Gaussian Mixture Model (GMM) is used to propose a biometric method based on heart sound signal.

METHODSFirstly, the original heart sounds signal was preprocessed by wavelet denoising. Then, Linear Prediction Cepstral Coefficients (LPCC) and Mel Frequency Cepstral Coefficients (MFCC) are compared to extract representative features and develops hidden Markov model (HMM) for signal classification. At last, the experiment collects 100 heart sounds from 50 people to test the proposed algorithm.

RESULTSThe comparative experiments prove that LPCC is more suitable than MFCC for heart sound biometric, and by wavelet denoising in each piece of heart sound signal, the system achieves higher recognition rate than traditional GMM.

CONCLUSIONThose results show that this method can effectively improve the recognition performance of the system and achieve a satisfactory effect.

Algorithms ; Biometry ; Heart ; physiology ; Humans ; Markov Chains ; Models, Biological ; Phonocardiography ; methods ; Wavelet Analysis