Objective: To address the challenges of systematically identifying herb pairs in traditional Chinese medicine (TCM), we proposed HerbGL, a framework for predicting potential herb pairs that integrates network propagation and graph regularization. Methods: Based on the assumption that herbal actions induce subtle perturbations in biological systems, a framework named HerbGL was proposed. Random walk with restart (RWR) was first applied to the protein-protein interaction (PPI) network to reconstruct herb-specific perturbation effects and generate weighted subnetworks. Then, to quantify affinity between herb pairs, two network-proximity metrics, Closeness and PageRank, were computed from the weighted subnetworks to construct herb-pair affinity matrices. Finally, these matrices, together with known herb pairs (derived from co-occurrence analysis of TCM formulas with a threshold determined from the inflection point of the frequency distribution), were incorporated into a graph regularization model to predict potential herb pairs. Model performance was assessed through baseline comparison, ablation and robustness experiment under different ratios of positive and negative samples, using the area under the receiver operating characteristic curve (AUROC), the area under the precision-recall curve (AUPRC), accuracy, and precision as evaluation metrics. Furthermore, the predicted herb pairs were validated through both literature evidence and Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses. Results: The weighted subnetworks constructed by RWR provided a refined simulation of herb-specific perturbation effects, which formed the basis for subsequent affinity modeling and prediction. Analysis of herb pair co-occurrence frequencies revealed a marked change around 150, which was selected as the threshold to distinguish herb pairs from non-herb pairs. HerbGL exhibited superior predictive performance compared with baseline models (AUROC = 0.970 5, AUPRC = 0.955 5, accuracy = 0.726 6, precision = 0.970 6). Ablation results showed that removing the Closeness and PageRank metrics substantially degraded performance (AUROC = 0.819 1, AUPRC = 0.876 8), confirming their necessity. Robustness evaluation under an imbalanced positive-to-negative sample ratio of 1 : 5 yielded AUROC = 0.969 6 and AUPRC = 0.840 4, indicating stable predictive ability. Moreover, multiple case studies further validated the rationality of the predicted herb pairs, such as Fangfeng (Saposhnikoviae Radix) and Qingpi (Citri Reticulatae Pericarpium Viride) which are recorded in Liangpeng Huiji (《良朋汇集》, Collection of Excellent Recipes) Vol. 3: Fangfeng Shengma Tang (防风升麻汤). Additionally, pathway enrichment analysis of the Renshen (Ginseng Radix et Rhizoma) and Lianqiao (Forsythiae Fructus) pair further supported the biological plausibility of their compatibility. Conclusion HerbGL offers an effective and biologically informed framework for identifying herb pairs in TCM. Beyond improving herb pair prediction, the framework also provides data support for research on herb compounds and mechanisms, thereby supporting data-driven exploration of TCM compatibility.

Objective:To explore the application value of part-cut jejunal transection in digestive tract reconstruction of totally laparoscopic total gastrectomy.Methods:The propensity score matching and retrospective cohort study was conducted. The clinicopathological data of 112 patients with gastric cancer who underwent totally laparoscopic total gastrectomy in The Second Affiliated Hospital of Anhui Medical University from June 2018 to September 2022 were collected. There were 81 males and 31 females, aged (70±8)years. Among the 112 patients, 60 patients undergoing diges-tive tract reconstruction by Roux-en-Y anastomosis with part-cut jejunum were set as the part-cut group, and 52 patients undergoing digestive tract reconstruction by traditional Roux-en-Y anasto-mosis were set as the traditional group. Observation indicators: (1) propensity score matching status and comparison of clinical data of patients between the two groups after matching; (2) intraopera-tive and postoperative conditions; (3) follow-up. Comparison of measurement data with normal dis-tribution between groups was conducted using the independent sample t test. Comparison of count data between groups was conducted using the chi-square test or Fisher exact probability. Com-parison of ordinal data was conducted using the nonparametic rank sum test. Propensity score matching was performed using the 1∶1 nearest neighbor matching method, with the caliper value of 0.02. Results:(1) Propensity score matching status and comparison of clinical data of patients between the two groups after matching. Of the 112 patients, 90 patients were successfully matched, with 45 cases in each of the part-cut group and the traditional group. After propensity score matching, the elimination of body mass index, clinical TNM staging confounding bias ensured comparability. (2) Intraoperative and postoperative conditions. After propensity score matching, the total operation time and digestive tract reconstruction time of patients in the part-cut group were (217.0±15.1)minutes and (34.7±1.8)minutes, versus (252.6±21.9)minutes and (52.6±7.4)minutes in the traditional group, respectively, showing significant differences in the above indicators between the two groups ( t=?8.97, ?15.66, P<0.05). (3) Follow-up. After propensity score matching, 90 patients were followed up postoperatively for (47±15)months. During the follow-up, no patient in either group received secondary surgery, and there was no death. There were 3 cases and 10 cases of Roux stasis syndrome in the part-cut group and the traditional group, respectively, showing a significant difference between the two groups ( χ2=4.41, P<0.05). Conclusion:Compared with traditional Roux-en-Y anastomosis, the Roux-en-Y anastomosis with part-cut jejunum in totally laparoscopic total gastrectomy can signifi-cantly shorten the time for digestive tract reconstruction and reduce the incidence of postoperative Roux stasis syndrome.

Objective To explore the mechanism of Qingda Granules(QDG)for alleviating brain damage in spontaneously hypertensive rats(SHRs).Methods Twelve 5-week-old SHRs were randomized into SHR control group and SHR+QDG group treated with QDG by gavage at the daily dose of 0.9 g/kg for 12 weeks.The control rats,along with 6 age-matched WKY rats,were treated with saline only.Blood pressure changes of the rats were monitored,and pathologies and neuronal apoptosis in the cerebral cortex were examined with HE staining and TUNEL staining.Cerebral cortical expressions of miR-124 and STAT3 mRNA were detected using RT-qPCR,and the protein expressions of NeuN,STAT3,Bcl-2,Bax,and cleaved caspase-3 were detected with immunohistochemistry and Western blotting.In a HT22 cell model of oxygen and glucose deprivation/reoxygenation(OGD/R),the effects of QDG on cell viability and apoptosis,expressions of miR-124 and STAT3 mRNA,and protein expressions of STAT3,Bcl-2,Bax,and cleaved caspase-3 were evaluated using CCK8 assay,Hoechst 33342 staining,RT-qPCR,and Western blotting.Results Compared with WKY rats,SHRs had significantly elevated systolic blood pressure,diastolic blood pressure and mean arterial pressure with significantly increased neuronal apoptosis in the cerebral cortex,reduced expressions of NeuN,miR-124 and Bcl-2,and enhanced expressions of STAT3,Bax and cleaved caspase-3(P＜0.05).All these changes in the SHRs were significantly ameliorated by treatment with QDG(P＜0.05).In the HT22 cell model,QDG treatment obviously reduced OGD/R-induced cell apoptosis,increased the expressions of miR-124 and Bcl-2,and suppressed the elevation of protein expressions of STAT3,Bax and cleaved caspase-3.Conclusion QDG inhibits cerebral cortical neuronal apoptosis and thereby attenuates brain damage in SHR rats by modulating the miR-124/STAT3 signaling axis.

Objective:To analyze the applied effect of artificial intelligence(AI)rehabilitation system combined with telemedicine platform in individualized rehabilitation treatment for patients with long-term coma,so as to realize individual management for them.Methods:An AI rehabilitation system that aimed to patients with long-term coma was designed by integrating data of multimodal sensors included inertial sensor,image sensor and bioelectric sensor.A telemedicine platform,which included seven modules(patient management,consultation management,video communication,data transmission,electronic medical records,cost management and system maintenance),was designed by using a browser/server(B/S)architecture.The changes of patients'condition were followed up,and the full process of the management for rehabilitation treatment and the individualization of treatment plan were realized.A total of 60 inpatients whose coma duration exceeded 28 days were selected from the Coma Awakening Center of Dongguan Shipai Hospital between October 1,2021 and September 30,2022.They were randomly divided into control group(n=30)and observation group(n=30)by random number table.The control group received physical therapy and rehabilitation training based on evaluation results of rehabilitation,while the observation group adopted AI rehabilitation system combined with the telemedicine platform to conduct intervention.The consciousness levels,quality of life,hospital stay duration and treatment costs of two group were compared after intervention.Results:The Glasgow Coma Scale(GCS)scores,Coma Recovery Scale-Revised(CRS-R)scores,and Persistent Vegetative State(PVS)scores of two groups after intervention were significantly higher than those before intervention,and these indicators of observation group were significantly higher than them of control group after intervention,and the differences were significant(t=7.187,15.586,9.293,P<0.05),respectively.There were not significant differences in Glasgow Outcome Scale(GOS)scores and Activities of Daily Living(ADL)scores between two groups before intervention(P>0.05),and these scores of two groups were significantly increased after intervention,and these scores of observation group were significantly higher than them of control group after intervention,and the differences were significant(t=7.584,6.755,P<0.05).Additionally,the hospital stay duration and treatment costs of observation group were significantly less than those of control group,and the differences were significant(t=20.965,9.503,P<0.05).Conclusion:The intervention of AI rehabilitation system combines with a telemedicine platform on patients with long-term coma can enhance GCS,CRS-R,PVS,GOS and ADL scores,and reduce hospital stay duration and treatment cost.

Objective:To explore the application value of part-cut jejunal transection in digestive tract reconstruction of totally laparoscopic total gastrectomy.Methods:The propensity score matching and retrospective cohort study was conducted. The clinicopathological data of 112 patients with gastric cancer who underwent totally laparoscopic total gastrectomy in The Second Affiliated Hospital of Anhui Medical University from June 2018 to September 2022 were collected. There were 81 males and 31 females, aged (70±8)years. Among the 112 patients, 60 patients undergoing diges-tive tract reconstruction by Roux-en-Y anastomosis with part-cut jejunum were set as the part-cut group, and 52 patients undergoing digestive tract reconstruction by traditional Roux-en-Y anasto-mosis were set as the traditional group. Observation indicators: (1) propensity score matching status and comparison of clinical data of patients between the two groups after matching; (2) intraopera-tive and postoperative conditions; (3) follow-up. Comparison of measurement data with normal dis-tribution between groups was conducted using the independent sample t test. Comparison of count data between groups was conducted using the chi-square test or Fisher exact probability. Com-parison of ordinal data was conducted using the nonparametic rank sum test. Propensity score matching was performed using the 1∶1 nearest neighbor matching method, with the caliper value of 0.02. Results:(1) Propensity score matching status and comparison of clinical data of patients between the two groups after matching. Of the 112 patients, 90 patients were successfully matched, with 45 cases in each of the part-cut group and the traditional group. After propensity score matching, the elimination of body mass index, clinical TNM staging confounding bias ensured comparability. (2) Intraoperative and postoperative conditions. After propensity score matching, the total operation time and digestive tract reconstruction time of patients in the part-cut group were (217.0±15.1)minutes and (34.7±1.8)minutes, versus (252.6±21.9)minutes and (52.6±7.4)minutes in the traditional group, respectively, showing significant differences in the above indicators between the two groups ( t=?8.97, ?15.66, P<0.05). (3) Follow-up. After propensity score matching, 90 patients were followed up postoperatively for (47±15)months. During the follow-up, no patient in either group received secondary surgery, and there was no death. There were 3 cases and 10 cases of Roux stasis syndrome in the part-cut group and the traditional group, respectively, showing a significant difference between the two groups ( χ2=4.41, P<0.05). Conclusion:Compared with traditional Roux-en-Y anastomosis, the Roux-en-Y anastomosis with part-cut jejunum in totally laparoscopic total gastrectomy can signifi-cantly shorten the time for digestive tract reconstruction and reduce the incidence of postoperative Roux stasis syndrome.

OBJECTIVES: To explore the mechanism of Qingda Granules (QDG) for alleviating brain damage in spontaneously hypertensive rats (SHRs). METHODS: Twelve 5-week-old SHRs were randomized into SHR control group and SHR+QDG group treated with QDG by gavage at the daily dose of 0.9 g/kg for 12 weeks. The control rats, along with 6 age-matched WKY rats, were treated with saline only. Blood pressure changes of the rats were monitored, and pathologies and neuronal apoptosis in the cerebral cortex were examined with HE staining and TUNEL staining. Cerebral cortical expressions of miR-124 and STAT3 mRNA were detected using RT-qPCR, and the protein expressions of NeuN, STAT3, Bcl-2, Bax, and cleaved caspase-3 were detected with immunohistochemistry and Western blotting. In a HT22 cell model of oxygen and glucose deprivation/reoxygenation (OGD/R), the effects of QDG on cell viability and apoptosis, expressions of miR-124 and STAT3 mRNA, and protein expressions of STAT3, Bcl-2, Bax, and cleaved caspase-3 were evaluated using CCK8 assay, Hoechst 33342 staining, RT-qPCR, and Western blotting. RESULTS: Compared with WKY rats, SHRs had significantly elevated systolic blood pressure, diastolic blood pressure and mean arterial pressure with significantly increased neuronal apoptosis in the cerebral cortex, reduced expressions of NeuN, miR-124 and Bcl-2, and enhanced expressions of STAT3, Bax and cleaved caspase-3 (P<0.05). All these changes in the SHRs were significantly ameliorated by treatment with QDG (P<0.05). In the HT22 cell model, QDG treatment obviously reduced OGD/R-induced cell apoptosis, increased the expressions of miR-124 and Bcl-2, and suppressed the elevation of protein expressions of STAT3, Bax and cleaved caspase-3. CONCLUSIONS QDG inhibits cerebral cortical neuronal apoptosis and thereby attenuates brain damage in SHR rats by modulating the miR-124/STAT3 signaling axis.
Animals ; Rats, Inbred SHR ; MicroRNAs/metabolism* ; STAT3 Transcription Factor/metabolism* ; Signal Transduction/drug effects* ; Drugs, Chinese Herbal/pharmacology* ; Rats ; Apoptosis/drug effects* ; Rats, Inbred WKY ; Male ; Hypertension

Objective:To analyze the applied effect of artificial intelligence(AI)rehabilitation system combined with telemedicine platform in individualized rehabilitation treatment for patients with long-term coma,so as to realize individual management for them.Methods:An AI rehabilitation system that aimed to patients with long-term coma was designed by integrating data of multimodal sensors included inertial sensor,image sensor and bioelectric sensor.A telemedicine platform,which included seven modules(patient management,consultation management,video communication,data transmission,electronic medical records,cost management and system maintenance),was designed by using a browser/server(B/S)architecture.The changes of patients'condition were followed up,and the full process of the management for rehabilitation treatment and the individualization of treatment plan were realized.A total of 60 inpatients whose coma duration exceeded 28 days were selected from the Coma Awakening Center of Dongguan Shipai Hospital between October 1,2021 and September 30,2022.They were randomly divided into control group(n=30)and observation group(n=30)by random number table.The control group received physical therapy and rehabilitation training based on evaluation results of rehabilitation,while the observation group adopted AI rehabilitation system combined with the telemedicine platform to conduct intervention.The consciousness levels,quality of life,hospital stay duration and treatment costs of two group were compared after intervention.Results:The Glasgow Coma Scale(GCS)scores,Coma Recovery Scale-Revised(CRS-R)scores,and Persistent Vegetative State(PVS)scores of two groups after intervention were significantly higher than those before intervention,and these indicators of observation group were significantly higher than them of control group after intervention,and the differences were significant(t=7.187,15.586,9.293,P<0.05),respectively.There were not significant differences in Glasgow Outcome Scale(GOS)scores and Activities of Daily Living(ADL)scores between two groups before intervention(P>0.05),and these scores of two groups were significantly increased after intervention,and these scores of observation group were significantly higher than them of control group after intervention,and the differences were significant(t=7.584,6.755,P<0.05).Additionally,the hospital stay duration and treatment costs of observation group were significantly less than those of control group,and the differences were significant(t=20.965,9.503,P<0.05).Conclusion:The intervention of AI rehabilitation system combines with a telemedicine platform on patients with long-term coma can enhance GCS,CRS-R,PVS,GOS and ADL scores,and reduce hospital stay duration and treatment cost.

Objective:To study the distribution of species type, biotype and genotype of human Brucella isolated and identified in Nanjing. Methods:A total of 89 strains of human Brucella were collected from microbiology laboratories of three sentinel hospitals in Nanjing from 2017 to 2022. The species type was identified using biological methods and Brucella nucleic acid detection (BCSP31-PCR and AMOS-PCR). Further biotyping of Brucella melitensis isolates was conducted by serological results of A and M factors. Meanwhile, genotype analysis was performed using multiple-locus variable number tandem repeat analysis (MLVA), multilocus sequence typing (MLST) and single nucleotide polymorphism (SNP). Results:From 2017 to 2022, 89 strains of Brucella isolated and identified in Nanjing were all Brucella melitensis. Among them, Brucella melitensis biotype 3 accounted for 82.02% (73/89), and biotype 1 accounted for 17.98% (16/89). MLVA typing showed that 89 strains of Brucella melitensis belong to the "Eastern Mediterranean" cluster and could be divided into 50 MLVA genotypes; among which panel 1 had 3 genotypes, namely Type 42 (84.27%, 75/89), Type 63 (8.99%, 8/89) and Type 43 (6.74%, 6/89). The results of MLST-9 and MLST-21 were both ST8, and the core genome multilocus sequence typing (cgMLST) classified 89 strains into 11 genotypes. SNP analysis revealed a total of 4 013 SNP loci, with SNPs ranging from 0 to 409 across different strains, involving 59 SNP genotypes. Conclusions:The human Brucella strains isolated and identified in Nanjing are all Brucella melitensis, mainly biotype 3. The MLVA cluster is the "Eastern Mediterranean" cluster. The traditional MLST-9 and MLST-21 typing results are all ST8 type, while cgMLST divides all the strains into 11 genotypes with higher resolution.

Based on the previous publications， it is believed that damp-heat syndrome is the core syndrome of rheumatoid arthritis （RA）， and Qingre Huoxue Formula （清热活血方） is an effective formula for the treatment of damp-heat syndrome of RA. “Inflammation-bone destruction” is a key pathological link of RA， and it is also the advantage of the effectiveness of Qingre Huoxue Formula. Leucine rich α-2-glycoprotein 1 （LRG1） can mediate the transforming growth factor-β （TGF-β） signalling pathway to participate in the pathogenic process of “inflammation-bone destruction” of RA， and it can be used as a target protein in the treatment of damp-heat syndrome of RA by Qingre Huoxue Formula. Accordingly， a scientific hypothesis was proposed that Qingre Huoxue Formula may regulate TGF-β signalling pathway mediated by LRG1 to improve “inflammation-bone destruction” of RA， and it was envisioned that the clinical effect of Qingre Huoxue Formula on LRG1 could be confirmed through clinical studies， and the mechanism of action of Qingre Huoxue Formula on the LRG1/TGF-β signalling axis as well as the influence of the expression or non-expression of the LRG1/TGF-β signalling axis on the therapeutic effectiveness of Qingre Huoxue Formula could be clarified through animal experiments.

OBJECTIVE To explore the construction of the physical identification model of twenty-five people of yin and yang by combining the hand diagnosis image analysis of machine learning,in order to enrich the input of physical identification features.METHODS Based on the theory of twenty-five people of yin and yang,the hand images of 542 patients with cardiovascular disease were collected by self-developed hand diagnosis acquisition equipment.The key points of the hand were extracted by the Mediapipe palm key point algorithm,and the palm area was segmented by geometry to obtain the length and color gamut characteristics of different parts.The physician with clinical license was used to determined the patient's yin and yang twenty-five constitution according to the e-lectronic medical record,and the patient's constitution was identified by comparing Random Forest,Logistic Regression,Xgboost algo-rithm and Light GBM.RESULTS This study successfully constructed a TCM constitution identification model based on hand charac-teristics,which can identify ten basic constitution types including gold wood,gold fire,gold water,gold soil,wood fire,wood water,wood soil,water fire,water soil and fire soil.After sample equalization,the Random Forest model performed best in physical identifi-cation,with an accuracy rate of 0.69,which was significantly better than other models.CONCLUSION The constitution identifica-tion model of TCM hand diagnosis based on hand characteristics can realize the constitution identification of the five elements of TCM.However,due to the use of only hand diagnosis data of TCM,the accuracy needs to be improved.Compared with the traditional body mass scale,it is more objective,convenient,highly stable and consistent.In the future,multimodal integration with face diagno-sis and whole body inspection can be carried out to promote the accuracy,intelligence and multi-dimensional evaluation ability of TCM constitution identification technology.