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.

OBJECTIVE To investigate the serum toll-like receptor 4(TLR4),matrix metalloproteinase-3(MMP-3),in-terleukin-1β(IL-1β)and peripheral blood helper T-cell 17/regulatory T-cell(Th17/Treg)levels in patients with chronic periodontitis infected with porphyromonas gingivalis(Pg),as well as their diagnostic value of Pg infection in patients with chronic periodontitis.METHODS Totally 390 cases of patients with chronic periodontitis who at-tended Tianjin Stomatological Hospital from Jan.2022 to Jan.2024 were selected as the study subjects,and were divided into the infected group(113 cases)and the uninfected group(277 cases)according to Pg infection,and those patients in the infected group were further classified into the mild group(31 cases),the moderate group(45 cases),and the severe group(37 cases)according to the condition of the disease.Serum TLR4,MMP-3,IL-1β,and peripheral blood Th17/Treg levels were compared between the infected and uninfected groups and patients with different conditions of chronic periodontitis Pg infection,and the infected group was included in the positive and the uninfected group was included in the negative,and the receiver's operating characteristics(ROC)curves were plotted to obtain the area under the curve(AUC)to analyze the diagnostic value of single and combined de-tection of serum TLR4,MMP-3,and IL-1β,Peripheral blood Th17/Treg for Pg infection in patients with chronic periodontitis.RESULTS Serum TLR4,MMP-3,IL-1β and peripheral blood Th17/Treg levels in the infected group were(20.41±5.12)ng/ml,(6.12±2.03)μg/L,(10.24±2.19)μg/L and(0.32±0.10),higher than those in the uninfected group(P＜0.05).Serum TLR4,MMP-3,IL-1β and peripheral blood Th17/Treg levels in the se-vere group were(22.58±6.10)ng/ml,(8.12±2.67)μg/L,(12.49±3.10)μg/L and(0.52±0.15),higher than those in the moderate group[(14.78±3.44)ng/ml,(6.15±2.07)μg/L,(7.23±1.99)μg/L and(0.22±0.07)]and mild group(P＜0.05),which in the moderate group was higher than those in the mild group(P＜0.05).The results of ROC analysis showed that the AUC of serum TLR4,MMP-3,IL-1β,and peripheral blood Th17/Treg combined test for diagnosing Pg infection in patients with chronic periodontitis was 0.937,which was higher than that of the four single tests(P＜0.05).CONCLUSION Serum TLR4,MMP-3,IL-1β and peripheral blood Th17/Treg were highly expressed in chronic periodontitis patients infected with Pg,and all four indicators can participate in the progression of the disease,and the combined detection of these four indicators has a higher diagnostic value for Pg infection in patients with chronic periodontitis.

Aim To investigate the expression of or-ganic anion transporting polypeptide 4C1(Oatp4c1)-P-glycoprotein(P-gp)in the kidneys of obese mice in-duced by high-fat diet(HFD),and its impact on the pharmacokinetic changes of digoxin.Methods C57BL/6 mice were randomly divided into the Chow group and the HFD group.Body weight and blood glu-cose were recorded weekly.After successful model es-tablishment,digoxin was intraperitoneally injected,and blood was collected at different time points.Part of the blood samples was used for LC-MS/MS detection,and the other part was used for the detection of other bio-chemical indicators.After 16 weeks,the organs were removed and weighed.HE and immunohistochemical staining was used to observe the renal pathology and the expression of Villin,a marker of proximal tubules.Western blot and qPCR were combined to detect the expression of Villin,Oatp4c1 and P-gp.Results In the HFD group,body weight and blood glucose in-creased significantly.The blood concentration of digox-in rose,the area under the curve increased,and the half-life was prolonged.The proximal tubular epithelial cells shed,and the protein expression of Villin,Oatp4c1 and P-gp decreased significantly.Conclu-sions The down-regulation of Oatp4c1-P-gp expres-sion in the kidneys of HFD mice leads to an increase in the blood concentration of digoxin and a decrease in re-nal clearance.

Objective Human facial morphology is an appearance phenotype with high heritability,high diversity,and complexity.Traditional facial morphological genetic analysis is mostly based on facial landmark measurements,using linear regression for genome-wide association studies,but this method extracts limited facial morphological feature information.This study established an extraction method for multidimensional facial representations and validated the correlation between 473 single-nucleotide polymorphisms(SNPs)previously reported to be significantly associated with facial features and facial representations in the Han Chinese population.Methods After acquiring facial 3D images,3D morphable face models and HR-net network were used to align and quantify the 3D images,obtaining high-density 3D facial point cloud data.After unsupervised clustering of the point cloud,principal component analysis was applied to reduce dimensionality and extract multidimensional morphological phenotypes for each facial region.Based on these multidimensional phenotypes,partial least squares regression(PLSR)and canonical correlation analysis(CCA)were used for genetic association analysis.Results A total of 10 SNPs were validated to be significantly associated with facial morphology in Han Chinese,of which 7 SNPs were validated by the PLSR method,2 SNPs were validated by the CCA method,and 1 SNP was validated by both methods.Conclusion Among the 10 significantly associated SNP sites,9 related facial morphological regions were consistent with previous reports in other populations,indicating that genes affecting complex facial morphology have cross-population effects.

OBJECTIVE To investigate the serum toll-like receptor 4(TLR4),matrix metalloproteinase-3(MMP-3),in-terleukin-1β(IL-1β)and peripheral blood helper T-cell 17/regulatory T-cell(Th17/Treg)levels in patients with chronic periodontitis infected with porphyromonas gingivalis(Pg),as well as their diagnostic value of Pg infection in patients with chronic periodontitis.METHODS Totally 390 cases of patients with chronic periodontitis who at-tended Tianjin Stomatological Hospital from Jan.2022 to Jan.2024 were selected as the study subjects,and were divided into the infected group(113 cases)and the uninfected group(277 cases)according to Pg infection,and those patients in the infected group were further classified into the mild group(31 cases),the moderate group(45 cases),and the severe group(37 cases)according to the condition of the disease.Serum TLR4,MMP-3,IL-1β,and peripheral blood Th17/Treg levels were compared between the infected and uninfected groups and patients with different conditions of chronic periodontitis Pg infection,and the infected group was included in the positive and the uninfected group was included in the negative,and the receiver's operating characteristics(ROC)curves were plotted to obtain the area under the curve(AUC)to analyze the diagnostic value of single and combined de-tection of serum TLR4,MMP-3,and IL-1β,Peripheral blood Th17/Treg for Pg infection in patients with chronic periodontitis.RESULTS Serum TLR4,MMP-3,IL-1β and peripheral blood Th17/Treg levels in the infected group were(20.41±5.12)ng/ml,(6.12±2.03)μg/L,(10.24±2.19)μg/L and(0.32±0.10),higher than those in the uninfected group(P＜0.05).Serum TLR4,MMP-3,IL-1β and peripheral blood Th17/Treg levels in the se-vere group were(22.58±6.10)ng/ml,(8.12±2.67)μg/L,(12.49±3.10)μg/L and(0.52±0.15),higher than those in the moderate group[(14.78±3.44)ng/ml,(6.15±2.07)μg/L,(7.23±1.99)μg/L and(0.22±0.07)]and mild group(P＜0.05),which in the moderate group was higher than those in the mild group(P＜0.05).The results of ROC analysis showed that the AUC of serum TLR4,MMP-3,IL-1β,and peripheral blood Th17/Treg combined test for diagnosing Pg infection in patients with chronic periodontitis was 0.937,which was higher than that of the four single tests(P＜0.05).CONCLUSION Serum TLR4,MMP-3,IL-1β and peripheral blood Th17/Treg were highly expressed in chronic periodontitis patients infected with Pg,and all four indicators can participate in the progression of the disease,and the combined detection of these four indicators has a higher diagnostic value for Pg infection in patients with chronic periodontitis.

Aim To investigate the expression of or-ganic anion transporting polypeptide 4C1(Oatp4c1)-P-glycoprotein(P-gp)in the kidneys of obese mice in-duced by high-fat diet(HFD),and its impact on the pharmacokinetic changes of digoxin.Methods C57BL/6 mice were randomly divided into the Chow group and the HFD group.Body weight and blood glu-cose were recorded weekly.After successful model es-tablishment,digoxin was intraperitoneally injected,and blood was collected at different time points.Part of the blood samples was used for LC-MS/MS detection,and the other part was used for the detection of other bio-chemical indicators.After 16 weeks,the organs were removed and weighed.HE and immunohistochemical staining was used to observe the renal pathology and the expression of Villin,a marker of proximal tubules.Western blot and qPCR were combined to detect the expression of Villin,Oatp4c1 and P-gp.Results In the HFD group,body weight and blood glucose in-creased significantly.The blood concentration of digox-in rose,the area under the curve increased,and the half-life was prolonged.The proximal tubular epithelial cells shed,and the protein expression of Villin,Oatp4c1 and P-gp decreased significantly.Conclu-sions The down-regulation of Oatp4c1-P-gp expres-sion in the kidneys of HFD mice leads to an increase in the blood concentration of digoxin and a decrease in re-nal clearance.

Aim To investigate the antagonistic effect of Schizandrin A(SchA)on oxidative stress-induced endothelial dysfunction and its mechanism of action.Methods Human umbilical vein endothelial cells(HUVECs)were selected as the research subjects,and the effects of SchA on cell viability were detected by MTT assay;the content of ROS in the cells was detec-ted by flow cytometry;the content of MDA and CAT in the cells,and the content of NO and ET-1 in the cell supernatant were detected by kit assay;and the expres-sion of SOD1,p-eNOS/eNOS proteins,and ET-1 in the cell supernatant were detected by Western blot.Immu-nofluorescence experiments were performed to detect Nrf2 entry into the nucleus of cells.Results SchA re-versed the LPS-or hypoxia-induced increase in ROS and MDA content as well as the decrease in SOD1 and CAT content in HUVECs by activating the Nrf2/Keap1/HO-1 signaling pathway.SchA inhibited the decrease of p-eNOS and eNOS protein expression in HUVECs cells,as well as NO content in cell culture medium and the increase of ET-1 content induced by LPS.The Nrf2 inhibitor ML385 reversed the antagonis-tic effects of SchA on oxidative stress and endothelial dysfunction.Conclusions SchA antagonized LPS and hypoxia-induced oxidative stress,and SchA amelio-rated oxidative stress-induced endothelial dysfunction by up-regulating the Nrf2/Keap1/HO-1 signaling path-way.

Objective Human facial morphology is an appearance phenotype with high heritability,high diversity,and complexity.Traditional facial morphological genetic analysis is mostly based on facial landmark measurements,using linear regression for genome-wide association studies,but this method extracts limited facial morphological feature information.This study established an extraction method for multidimensional facial representations and validated the correlation between 473 single-nucleotide polymorphisms(SNPs)previously reported to be significantly associated with facial features and facial representations in the Han Chinese population.Methods After acquiring facial 3D images,3D morphable face models and HR-net network were used to align and quantify the 3D images,obtaining high-density 3D facial point cloud data.After unsupervised clustering of the point cloud,principal component analysis was applied to reduce dimensionality and extract multidimensional morphological phenotypes for each facial region.Based on these multidimensional phenotypes,partial least squares regression(PLSR)and canonical correlation analysis(CCA)were used for genetic association analysis.Results A total of 10 SNPs were validated to be significantly associated with facial morphology in Han Chinese,of which 7 SNPs were validated by the PLSR method,2 SNPs were validated by the CCA method,and 1 SNP was validated by both methods.Conclusion Among the 10 significantly associated SNP sites,9 related facial morphological regions were consistent with previous reports in other populations,indicating that genes affecting complex facial morphology have cross-population effects.

Objective:To investigate the efficacy of radiotherapy in patients with advanced esophageal cancer receiving first-line chemoimmunotherapy.Methods:A retrospective analysis was conducted on the data of 137 patients with Stage Ⅳ esophageal squamous cell carcinoma (ESCC) treated at our hospital from January 2018 to May 2023. These patients were divided into two groups: a group treated with first-line chemoimmunotherapy combined with radiotherapy (chemoimmunotherapy + radiotherapy group, n = 43) and a group treated with only chemoimmunotherapy ( n = 94). Inverse probability of treatment weighting (IPTW) was applied to balance baseline characteristics between the groups. With overall survival (OS) and progression-free survival (PFS) as study endpoints, the survival data were analyzed using the Kaplan-Meier method, the log-rank test, and the Cox regression method. Results:Before calibration, the chemoimmunotherapy + radiotherapy group significantly outperformed the sole chemoimmunotherapy group in median PFS (13.6 months vs. 7.0 months; HR: 0.501, 95% CI: 0.309-0.811, P = 0.005). After calibration using the COX proportional-hazards model for age, gender, Eastern Cooperative Oncology Group (ECOG) performance status, smoking history, T/N/M stage, and tumor location, the chemoimmunotherapy + radiotherapy group still had significant advantages in PFS (14.7 months vs. 7.0 months; HR: 0.441, 95% CI: 0.261-0.745, P = 0.002). IPTW analysis further confirmed this trend (13.9 months vs. 7.0 months; HR: 0.492, 95% CI: 0.304-0.795, P < 0.001). Specifically, the median OS of the chemoimmunotherapy + radiotherapy group demonstrated significant improvement in all analyses: pre-calibration (29.5 months vs. 18.0 months; HR: 0.507, 95% CI: 0.297-0.867, P = 0.013), after calibration using the Cox model (27.5 months vs. 16.7 months; HR: 0.470, 95% CI: 0.266-0.830, P = 0.009), and after calibration using IPTW (29.5 months vs. 16.9 months; HR: 0.448, 95% CI: 0.262-0.764, P < 0.001). Conclusions:The combination of radiotherapy and first-line chemoimmunotherapy can significantly improve survival outcomes of patients with advanced ESCC, suggesting its potential as a standard treatment strategy.

ObjectiveTo explore the potential mechanism of Huoxue Xiaoyi Formula （活血消异方， HXF） in treating ovarian endometriosis （OEM） from the perspective of T follicular helper （Tfh） cells and the Janus kinase/signal transducer and activator of transcription （JAK/STAT） signaling pathway. MethodsForty-five female SD rats with normal estrous cycles were randomly divided into three groups， HXF group， model group， and normal group， with 15 rats in each group. A rat model of OEM was established by autologous endometrial tissue implantation. After successful modeling， the treatment group received HXF at 5.85 g/（kg·d） by gavage for 14 consecutive days. The model group and normal group received 1 mL/d of normal saline by gavage. RNA-sequencing data from human proliferative-phase endometriotic and normal endometrial tissues were downloaded from the GEO database. Transcriptomic sequencing was used to analyze gene expression in rat ovarian ectopic tissues and normal uterine tissues， and comparisons were made with human data to verify JAK/STAT pathway activation in proliferative-phase ectopic tissues. Immunohistochemistry was used to detect the positive expression of CXC chemokine receptor 5 （CXCR5） and interleukin-21 （IL-21） in rat ovarian ectopic and normal uterine tissues. Western Blotting was performed to detect the protein levels of IL-21， IL-21 receptor （IL-21R）， Janus kinase 1 （JAK1）， signal transducer and activator of transcription 6 （STAT6）， and B-cell lymphoma 2 （Bcl-2）. Tfh cell infiltration was analyzed using immune cell infiltration methods. ResultsGene set enrichment analysis showed that the JAK/STAT pathway was significantly activated in human proliferative-phase endometriotic tissues compared to normal endometrial tissues. Similarly， the JAK/STAT pathway was markedly activated in rat ovarian ectopic tissues in the model group compared to the normal group， but suppressed in the HXF group compared to the model group. Compared with normal uterine tissues， ovarian ectopic tissues in the model group showed increased Tfh cell infiltration scores， higher CXCR5 and IL-21 expression， and elevated levels of IL-21， IL-21R， JAK1， STAT6， and Bcl-2 proteins. Compared with the model group， HXF group showed reduced CXCR5 and IL-21 expression and decreased protein levels of IL-21， IL-21R， JAK1， STAT6， and Bcl-2. ConclusionHXF may suppress activation of the JAK/STAT signaling pathway in ovarian endometriotic tissues by inhibiting IL-21 secretion from Tfh cells.