Intelligent prescription recommendation has become an important research direction in traditional Chinese medicine (TCM), offering new opportunities to support clinical decision-making and promote the modernization of TCM practice. With the rapid development of artificial intelligence (AI), a variety of computational approaches have been proposed to learn prescription patterns from clinical data and generate personalized treatment recommendations. However, despite increasing research activity, systematic and comprehensive reviews of AI-driven methods for TCM prescription recommendation remain limited. In this study, we present a comprehensive review of computational approaches for herbal prescription recommendation (HPR) in TCM. Existing methods are systematically categorized into several major paradigms, including traditional machine learning methods, topic model methods, sequential generative methods, deep learning and graph-based methods, and large language model–based frameworks. In addition to summarizing methodological developments, we also review commonly used public datasets and evaluation metrics in this field. Furthermore, representative models with publicly available implementations are experimentally evaluated on multiple benchmark datasets to provide a comparative analysis of their performance on the HPR task. Finally, we discuss the key challenges that hinder the practical deployment of intelligent prescription recommendation systems, including data heterogeneity, limited interpretability, and insufficient integration of TCM domain knowledge. Future research directions are outlined to facilitate the development of more reliable, interpretable, and clinically applicable AI-assisted HPR systems for TCM.

ObjectiveTo explore the potential mechanism of Changji'an Formula （肠激安方） on intestinal permeability for rats with diarrhea-predominant irritable bowel syndrome （IBS-D） of liver depression and spleen deficiency syndrome by the microRNA-29b-3p （miR-29b-3p）/tumor necrosis factor receptor-associated factor 3 （TRAF3）/nuclear factor-κB （NF-κB）/myosin light chain kinase （MLCK） axis. MethodsTwenty-four 1-day-old male Sprague-Dawley （SD） suckling rats were selected， and the IBS-D rat model of liver depression and spleen deficiency syndrome was established via a three-factor method，i.e. maternal separation plus acetic acid stimulation and restraint stress， for 6 consecutive weeks. After successful modeling， the rats were randomly divided into a model group， pinaverium bromide group， low-dose and high-dose Changji'an Formula groups， with 6 rats in each group. Another 6 age-matched non-modeled SD rats were included as the control group. The low-dose and high-dose Changji'an Formula groups were given intragastric administration of Changji'an Formula solution at doses of 16.74 g/（kg·d） and 33.48 g/（kg·d）， respectively； the pinaverium bromide group received intragastric administration of pinaverium bromide tablets at 0.018 g/（kg·d）； and the control group was given distilled water at 10 ml/（kg·d） via intragastric gavage. The intervention was conducted once daily for 14 consecutive days. After the gavage treatment， the fecal water content of rats in each group was measured. Enzyme-linked immunosorbent assay （ELISA） was used to detect the serum levels of intestinal permeability indicators， including D-lactic acid （D-LA）， diamine oxidase （DAO）， and lipopolysaccharide （LPS）. Quantitative real-time polymerase chain reaction （qRT-PCR） was conducted to determine the mRNA expression levels of miR-29b-3p， TRAF3， tumor necrosis factor-α （TNF-α）， p65， p50， and MLCK in colonic tissues. Western Blot analysis was employed to detect the protein expression levels of TRAF3， TNF-α， p65， phosphorylated p65 （p-p65）， MLCK， myosin light chain （MLC）， phosphorylated MLC （p-MLC）， and tight junction proteins including junctional adhesion molecule-A （JAM-A）， Occludin， and Claudin-1 in colonic tissues. ResultsCompared with the control group， the model group exhibited significantly increased fecal water content and serum levels of D-LA， DAO， and LPS， along with decreased protein expression levels of JAM-A， Occludin， and Claudin-1 in colonic tissues （P＜0.05 or P＜0.01）. Additionally， in the model group， the mRNA expression levels of miR-29b-3p， TNF-α， p65， p50， and MLCK in colonic tissues were up-regulated， while the mRNA and protein expression levels of TRAF3 were down-regulated； the protein levels of TNF-α and MLCK， as well as the ratios of p-p65/p65 and p-MLC/MLC， significantly elevated （P＜0.01）. Compared with the model group， all treatment groups showed reduced fecal water content and serum levels of D-LA， DAO， and LPS， along with down-regulated mRNA expression levels of miR-29b-3p， TNF-α， p65， p50， and MLCK， and up-regulated TRAF3 mRNA expression in colonic tissues. Moreover， the pinaverium bromide group and high-dose Changji'an Formula group presented increased protein levels of Occludin， Claudin-1， and TRAF3， as well as decreased protein levels of TNF-α and MLCK， and reduced ratios of p-p65/p65 and p-MLC/MLC in colonic tissues （P＜0.05 or P＜0.01）. Compared with the low-dose Changji'an Formula group， the high-dose group had lower fecal water content and serum levels of DAO and LPS （P＜0.01）. In comparison with the pinaverium bromide group， the high-dose Changji'an Formula group showed a significant decrease in serum DAO level （P＜0.01）. ConclusionsChangji'an Formula can reduce intestinal permeability and restore intestinal barrier function in IBS-D rats of liver depression and spleen deficiency syndrome by regulating the miR-29b-3p/TRAF3/NF-κB/MLCK axis.

Objective The study aimed to elucidate the population composition and seasonal distribution characteristics of mosquitoes in Laoshan District, Qingdao City, providing a scientific basis for the prevention and control of mosquito-borne infectious diseases. Methods Mosquito surveillance was conducted in Laoshan District in Qingdao City using light traps from April to November 2020-2022. Dominant mosquito species were identified using the Berger-Parker dominance index(I)and mosquito density was compared using the Kruskal-Wallis H test. The seasonal distribution characteristics of adult mosquitoes were analyzed using concentration and circular distribution method. Spearman rank correlation analysis was used to study the relationship between mosquito populations and meteorological factors with mean monthly temperature, mean monthly relative humidity, and mean monthly rainfall. Results The mean mosquito density for the trapping period was 9.771 females/(trap·night)in 2020,9.771 in 2021, and 9.427 in 2022. There was no statistically significant difference in mosquito density between different years(H = 0.095, P = 0.954). In total, 2 781 adult female mosquitoes comprising five species from four genera were captured over three years. Culex pipiens(I = 0.654), Aedes albopictus(I = 0.202), and Aedes aegypti(I = 0.118)were identified as dominant species. Seasonal fluctuations in the adult mosquitoes were unimodal, and primarily concentrated from June to September. The peak period of adult mosquito varied between years with statistical significance. (F = 3.838, P ＜ 0.05). Mosquito density was highly correlated with mean monthly temperature, mean monthly relative humidity and mean monthly precipitation(P＜ 0.05). Conclusions Cx. pipiens pallens, Ae. albopictus, and Ar. inamoratus are the dominant mosquito species in the Laoshan District of Qingdao City. The activity of adult mosquitoes is seasonal, and primarily concentrated in summer and autumn, during which the risk of mosquito-borne diseases such as dengue fever is increased.

Objective To investigate the capacity of municipal and county-level Centers for Disease Control (CDC) in Shandong Province for γ radionuclides in food as well as gross α and gross β in drinking water, and to provide evidence for strengthening radiological laboratories. Method Data regarding radioactive detection capacity were collected from 16 prefecture-level cities and 136 county-level CDCs. Chi-square test and Fisher's exact tests were used to compare capacity differences across regions and between cities with and without nuclear power plants. Correlation analysis was used to evaluate the spatial equity of gross α and gross β detection capacity for drinking water. Result The detection capacity of gross α and gross β in drinking water varied significantly among counties (χ²=81.026, P<0.001），no significant difference was found between coastal and inland cities (χ²=2.021，P>0.155); Cities with nuclear power plants showed better detection capacity (χ²=20.253，P<0.01). The detection capacity for γ radionuclides in food was primarily concentrated in the prefecture-level cities that host nuclear power plants. Conclusion The overall detection capacity for radioactivity in drinking water was acceptable but spatially uneven. The capacity for monitoring radioactive contamination in food was generally weak. It is necessary to strengthen capacity building at the grassroots level.

Gastric cancer is one of the most common malignant tumors in the digestive system, characterized by high incidence and mortality rates. In recent years, with the rapid develop-ment of molecular immunology, the application of immune checkpoint inhibitors (ICIs) in neoadju-vant therapy has significantly improved pathological response rates and survival outcomes for patients with resectable locally advanced gastric cancer. The authors systematically review current research progress on combination strategies involving immune checkpoint inhibitors in neoadjuvant therapy for locally advanced gastric cancer, aiming to provide an evidence for optimizing individua-lized therapeutic regimens.

Objective To study the effect of peripheral blood CD4+T lymphocytes of syphilis patients on human immunodeficiency virus type 1(HIV-1)infection and replication,and explore the molecular mechanism of Trepone-ma pallidum(TP)in facilitating HIV-1 replication.Methods Peripheral blood mononuclear cells(PBMCs)were isolated from syphilis patients and TP-uninfected healthy blood donors,then infected with HIV-1 in vitro.HIV-1 replication level was studied.PBMCs of healthy persons were stimulated with TP and then infected with HIV-1.CD4+T lymphocytes were isolated using flow cytometry,and changes in host gene expression profiles were verified using messenger RNA(mRNA)high-throughput sequencing and real-time fluorescence quantitative polymerase chain reaction(RT-qPCR)methods.Results CD4+T lymphocytes from TP-infected persons and TP-stimulated CD4+T lymphocytes from healthy persons could significantly enhance HIV-1 infection and viral replication levels.The HIV-1 p24 content in the supernatant of cultured cells and intracellular HIV-1 DNA copy number were both higher than those in the control group(both P＜0.05).There were significant differences in the mRNA gene ex-pression profiles between the TP stimulation combined with HIV-1 infection group and the HIV-1 infection alone group.Among them,the expression levels of chemokines CXCL5 and CXCL8 were significantly upregulated in the TP stimulation combined with HIV-1 infection group,indicating that chemokines might play an important role in the enhancement of HIV-1 infection and replication by TP.Conclusion TP infection can significantly enhance the sus-ceptibility of peripheral blood CD4+T lymphocytes to HIV-1 and promote virus replication,providing new scientific evidences for TP infection promoting HIV-1 infection.

Objective:To investigate the efficacy and safety of sintilimab combined with the SOX regimen for adjuvant treatment of stage Ⅲgastric cancer after D2 radical resection and to provide a reference for individualized clinical treatment.Methods:The clinical data of 245 pa-tients with stage III gastric cancer who underwent D2 radical resection at the 940th Hospital of the Joint Support Force of the People's Liber-ation Army from June 2019 to May 2022 were retrospectively analyzed.The 180 patients who received only the SOX regimen were desig-nated the control group,and the 65 patients who received sintilimab combined with the SOX regimen were designated the experimental group.The 3-year disease-free survival(DFS)rate,overall survival(OS)rate,and adverse reactions among the two groups and different sub-groups(HER-2 positive,dMMR,CPS≥5)were compared.Results:The 3-year DFS(81.5%vs.59.4%)and OS(84.6%vs.70.6%)rates in the experimental group were significantly higher than those in the control group(both P<0.05).Group analysis showed that in patients with CPS≥5,the 3-year DFS(91.5%vs.67.0%)and OS(95.7%vs.71.6%)rates within the experimental group were significantly better than those in the control group(both P<0.05).Intra-group analysis within the experimental group showed that the 3-year DFS rate(91.5%vs.55.6%)and OS rate(95.7%vs.55.6%)of patients with CPS≥5 were significantly better than those of patients with CPS<5(both P<0.05).The overall and grade≥3 incidences of liver and kidney function damage,thyroid dysfunction,colitis,pneumonia,and rash in the experimental group were higher than those in the control group(all P<0.05),while the differences in other adverse reactions,including leukopenia were not statistic-ally significant(all P>0.05).Conclusions:Sintilimab combined with the SOX regimen can significantly improve 3-year DFS and OS rates in pa-tients with stage Ⅲ gastric cancer after surgery,especially in the CPS≥5 subgroup,with significant benefits and controllable safety.

Objective:To use machine learning to predict the efficacy of accelerated corneal collagen cross-linking (A-CXL) surgery, identify prognostic factors, and construct models to predict postoperative disease progression.Methods:A single-center retrospective study was conducted.A total of 82 keratoconus patients (112 eyes) who underwent A-CXL surgery at the West China Hospital of Sichuan University between March and December 2021 were enrolled.Preoperative and follow-up examinations included anterior segment evaluation by slit-lamp microscopy, corneal topography using Pentacam, and corneal biomechanical indices using Corvis ST.Disease progression was defined as an increase in maximum keratometry (Kmax) of ≥1 D from the preoperative level at the last follow-up.Various machine learning algorithms were employed to analyze corneal topography, biomechanical parameters and corneal densitometry values to identify prognostic factors and construct models for predicting postoperative disease progression.This study adhered to the Declaration of Helsinki.The study protocol was approved by the Ethics Committee of West China Hospital, Sichuan University (No.2023496).Written informed consent was obtained from each subject.Results:During follow-up, 15.1% (17/112) of the eyes showed progression after A-CXL.The preoperative astigmatism and stress-strain index (SSI) in the progression group were (-5.41±2.72)D and 1.41±0.78, respectively, which were significantly higher than (-3.30±2.54)D and 0.95±0.98 in the non-progression group ( t=2.80, 2.03; both P<0.05).Cox regression analysis identified preoperative astigmatism (hazard ratio [HR]=1.20), SSI (HR=1.10), and anterior corneal densitometry of 2-6 mm (CDA6) (HR=2.10) as significant risk factors for post-A-CXL progression.Among various machine learning models developed and validated, the area under the curve (AUC) values for logistic regression, multilayer perceptron (MLP) model, and random forest (RF) exceeded 0.700.For F1-score, the AUC values for logistic regression, MLP, and RF were 0.870, 0.880, and 0.880, respectively.The network structure of the visualized MLP was a single-layer, 24-neurons neural network with 80% accuracy in predicting whether progression occurred after A-CXL.The clinical nomogram developed in conjunction with astigmatism, SSI, and CDA6 predicted the cumulative probability of progression at 0.5, 1, and 2 years postoperatively based on the sum of the specified values for each variable, and based on the optimal cutoff value, keratoconus corneas could be classified into high-, intermediate-, and low-risk groups, respectively.The time-dependent subject operating characteristic curves of the nomogram showed AUCs of 0.734, 0.685, and 0.935 at 0.5, 1, and 2 years postoperatively, respectively, all of which performed well in predicting progression. Conclusions:Preoperative astigmatism, SSI, and CDA6 are significant risk factors for post-A-CXL progression in keratoconus.The MLP model can accurately predict postoperative disease progression, and the clinical nomogram combining preoperative astigmatism, SSI, and CDA6 can effectively differentiate between low-, medium-, and high-risk postoperative progression outcomes.

Objective To design a performance testing platform to evaluate the working status and performance characteristics of the ventilator proportional solenoid valve.Methods The performance testing platform had its hardware including a high-pressure gas source,a pressure regulating valve,sensors and etc,and its software designed based on PyQt5 and composed of several modules for data acquisition,parameter setting,image display,indicator computation,result output and etc.Two kinds of proportional solenoid valves(Valve 1、Valve2)were selected for static and dynamic tests to verify the performance of the platform.Results The platform developed facilitated the proportional solenoid valve to carry out accurate computation of static and dynamic indicators at real time and time domain and waveform feature extraction of sensor data by precision control and data acquisition for the proportional solenoid valve.Static tests showed that Valve 1 gained advantages over Valve 2 in static flow characteristics involving in lowered repeatability,return error and offset while enhanced stability;dynamic tests indicated Valve 2 had rapid flow variations and significant flow fluctuation impacts,Valve 1 showed smooth dynamic response changes,and Valve 2 behaved better than Valve 1 in dynamic performance.Conclusion The testing platform developed comprehensively demonstrates the performance characteristics and working performance of the ventilator proportional solenoid valve,which is of great significance to enhance the reliability and safety of the ventilator.[Chinese Medical Equipment Journal,2025,46(1):13-19]

Objective To propose a classification model based on a pseudo-bag strategy and feature adjustment for whole slide imaging in pathology.Methods A pseudo-bag generator was constructed to divide a parent bag into 3 pseudo-bags for increasing the number of training bags.Then,a pseudo-bag learning method based on Nystr?m-based algorithm for approximating self-attention and a selective feature fusion method were employed to process the pseudo-bags.Specifically,the pseudo-bag learning method based on Nystr?m-based algorithm for approximating self-attention reduced computational complexity through an improved multi-head self-attention mechanism while deeply extracting instance features to obtain pseudo-bag classification predictions,thereby enhancing pseudo-bag classification accuracy;and the selective feature fusion method refined pseudo-bag features by filtering and extracting relevant instances.Finally,the model adjusted bag features by extracting confounding factors to avoid interference from irrelevant information and further improve classification accuracy.Results The proposed model was evaluated on two datasets(CAMELYON-16 and TCGA-NSCLC)and compared with 10 other methods,and the results demonstrated that the proposed model achieved the best performance.The proposed method reached an accuracy of 0.943 on the CAMELYON-16 dataset and 0.906 on the TCGA-NSCLC dataset.Conclusion The proposed model can significantly improve the accuracy of whole-slide pathological image classification by effectively mitigating the overfitting and avoiding interference from irrelevant information.