OBJECTIVE To construct a preoperative medication reconciliation model assisted by a localized large language model （LLM） for gastric cancer and evaluate its clinical efficacy. METHODS A total of 249 gastric cancer patients with a history of continuous medication before admission in the Gastric Surgery Department of Jiangsu Cancer Hospital were retrospectively enrolled. Patients were divided into training set （154 cases） and validation set （95 cases） based on the order of time. Based on guidelines， drug package inserts， and other evidence， a standardized medication reconcili ation process and a structured knowledge base were constructed. DeepSeek-V3 LLM was deployed privately in the hospital， combined with retrieval-augmented generation technology， to achieve automated integration of medication information， risk screening， and generation of personalized recommendations. The quality of LLM-generated recommendations was evaluated using automatic metrics （BERT Score and ROUGE-1， 2， L） and manual scoring [seven-dimensional index （7DI） ] . Spearman correlation analysis was performed to explore the correlation between automatic scores and manual scores. Cronbach’s α coefficient was used to test the internal consistency of manual scoring results. The time consumed by manual and LLM-assisted medication reconciliation was compared across tasks of different difficulty levels （simple， moderate， and high）. RESULTS A structured knowledge base covering 8 major drug categories was finally established， covering common and high-risk preoperative medication scenarios and providing structured retrieval support for the LLM. For automatic evaluation， the precision， recall， and F1-score of BERT Score were 0.783±0.033， 0.811±0.038， and 0.796±0.028， respectively. The F1-scores of ROUGE-1， ROUGE-2 and ROUGE-L were 0.566±0.067， 0.338±0.076 and 0.468±0.082， respectively. The 7DI scores from three manual raters ranged from 32.06 to 33.45. The F1-score of automatic scoring was significantly positively correlated with the 7DI score of manual scoring （maximum coefficient of determination＝0.611， P ＜0.001）， and the internal consistency of manual scoring was good （Cronbach’s α ＝ 0.876）. In terms of efficiency， LLM-assisted medication reconciliation reduced time consumption by more than 90% compared with manual reconciliation in the simple， moderate， and high-difficulty groups （ P ＜0.001）. CONCLUSIONS The medication reconciliation model constructed based on a localized LLM and structured knowledge base shows high accuracy， consistency， and clinical applicability in complex preoperative medication scenarios for gastric cancer. It can improve the efficiency of medication reconciliation and reduce potential medication risks.

ObjectiveTo investigate the effect of video-based educational intervention combined with maternal presence on perioperative adverse outcomes in preschool children undergoing general anesthesia， including cooperation in anesthesia induction， perioperative anxiety， pain and agitation during recovery. MethodsA total of 300 preschool children scheduled for general anesthesia in our hospital from June to December 2023 were randomly assigned to control group （n=150） and intervention group （n=150）. The control group received routine recovery care. For the intervention group， in addition to routine recovery care， a preoperative visit was scheduled one day before surgery. During this visit， mothers were guided to watch anesthesia videos with their children. During the waiting period in the operating room and 30 minutes after awakening， the mothers were guided to accompany the children for more than 30 minutes. Recovery conditions were recorded using the surgical anesthesia information system， and the children’s anesthetic induction compliance， perioperative anxiety， pain， and agitation were evaluated and recorded using the modified Yale Preoperative Anxiety Scale （m-YPAS）， the Induction Compliance Scale （ICC）， the Children’s Pain Behavior Scale （FLACC）， and the Pediatric Agitation and Emergence Delirium Scale （PAED）. ResultsOn the preoperative visit day， there were no statistically significant differences in baseline data between the two groups （P > 0.05）. For perioperative anxiety， the m-YPAS scores of the intervention group were significantly lower than those of the control group， both when entering the operating room waiting area （35.27±6.48 vs. 41.79±6.68， P < 0.05） and 30 minutes after postoperative recovery （20.13±7.05 vs. 35.75±9.51， P < 0.05）. In terms of anesthesia induction cooperation， the ICC scores of the intervention group were significantly lower than those of the control group （1.84±0.95 vs. 3.17±0.62， P < 0.05）， and the proportion of good induction cooperation was significantly higher than that of the control group （24.00% vs. 12.67%， P < 0.05）. There was no significant difference in awakening duration between the two groups， but the intervention group had a significantly shorter length of stay in the post-anesthesia care unit than the control group （0.90±0.29 hours vs. 1.29±0.42 hours， P < 0.001）. For perioperative agitation， the PAED scores of the intervention group were significantly lower than those of the control group （entering in the operating room waiting area： 8.5 vs. 9.2， P < 0.05； 30 minutes after postoperative recovery： 4.2 vs. 7.8， P < 0.05）. In terms of pain scores， the FLACC scores of the intervention group were also significantly lower than those of the control group， both when entering the operating room waiting area （ 5.3 vs. 6.7， P < 0.05； 30 minutes after postoperative recovery： 2.1 vs. 4.9， P < 0.05）. ConclusionsVideo-based educational intervention combined with maternal presence reduces the perioperative anxiety， pain and agitation of preschool children undergoing general anesthesia， and improved the compliance of anesthesia induction. It is recommended to promote this intervention measure in clinical practice.

Objective:To investigate the effects of intravenous lidocaine infusion (IVLI) on the quality of intraoperative neurophysiologic monitoring (IONM) in patients with thyroid tumor.Methods:A prospective randomized controlled study was conducted. A total of 60 patients with thyroid tumor undergoing thyroidectomy in the First Clinical Medical College of Shanxi Medical University between September 2022 and May 2023 were selected. According to the random number table method, all patients were divided into the lidocaine group and the control group, 30 cases in each group. The patients in the lidocaine group were continually given IVLI during the operation and the patients in the control group were continually given the equal 0.9% NaCl solution infusion during the operation. All patients in the 2 groups were induced by the same way of total intravenous anesthesia, and no muscle relaxants were added during anesthesia except the induction dose, and enhanced nerve monitoring tracheal catheter was inserted in the 2 groups. According to the standardized procedure of IONM, the electrode impedance values were measured at the time of eliciting the V1 and V2 signals from the vagus nerve, respectively, and the difference value was defined as the drop in the aggregate impedance level (DAIL). DAIL, perioperative hemodynamic parameters and postoperative recovery quality were compared between the 2 groups.Results:There were no statistically significant differences in the baseline data, operation time, intraoperative dosage of propofol and remifentanil between the 2 groups (all P > 0.05). Compared with the control group, the lidocaine group had a higher proportion of patients with DAIL<50% [80.0% (24/30) vs. 40.0% (12/30), χ2 = 10.00, P = 0.002], a lower hemodynamic fluctuation during extubation [mean arterial pressure: (95±6) mmHg (1 mmHg = 0.133 kPa) vs. (104±7) mmHg, t = 31.00, P < 0.001; heart rate: (73±5) times/min vs. (92±6) times/min, t = 172.58, P < 0.001], a lower visual analog score 24 h after surgery [(2.0±0.7) scores vs. (3.7±0.8) scores, t = -8.86, P < 0.001], a higher score of quality of recovery-15 scale [(127±11) points vs. (118±13) points, t = 2.92, P = 0.005]. Conclusions:IVIL can improve the quality of IONM in patients with thyroid tumor during surgery, reduce perioperative hemodynamic fluctuation and improve postoperative recovery quality of patients.

Objective To explore the effectiveness of"near peer learning"(NPL)in the electromyo-graphy(EMG)teaching module for neurology residents.Methods The Department of Neurology,Peking Union Medical College Hospital implemented an NPL instructional design for a course on EMG for residents from November 2020 to March 2024.This teaching session was held annually,in which senior residents in-structed juniors who were 1 or 2 years earlier in their training.The residents participated in the pre-course/post-course tests and completed a feedback survey at the end of the session.This evaluation method was used to un-derstand the effectiveness of the NPL intervention in EMG teaching.Results Over four years,a total of 83 residents participated.Among them,there were 24 postdoctoral students,52 postgraduates and 7 junior resi-dents.The results showed that the post-course test scores were significantly improved compared with pre-course test scores(74.33±2.43 vs.70.11±2.49,P=0.005),with the most remarkable improvements seen for"tu-tees"(73.84±20.53 vs.70.29±21.46,P=0.020),postgraduates(74.04±22.51 vs.68.97±21.40,P=0.009),first-year residents(70.19±4.02 vs.63.59±3.59,P=0.040)and first-time participating resi-dents(65.23±3.24 vs.60.97±3.21,P=0.030).The post-program feedback showed that both tutors and tu-tees thought highly of NPL,believing that it enabled them to gain knowledge and helped them to improve teaching skills.Conclusions The NPL intervention is suitable for the teaching of EMG,because of its contri-bution to knowledge acquisition and basic clinical skills improvement.The NPL is worth replicating in other teaching and learning programs.

Objective To explore the correlation between the expression level of serum fibrinogen-like protein 1(FGL1)and the indexes of metabolism and renal function in patients with diabetic nephropathy(DN)and diabetes mellitus(DM),and provide reference for clinical diagnosis and treatment.Methods From January 2017 to April 2023,30 patients with DM and treated in Jiangsu Province Hospital of Chinese Medicine were selected as the DM group,68 patients with DN were selected as the DN group,and 36 healthy subjects were selected as the control group.The DN group was further divided into the early DN(DN-E)group(n=38)and the late DN(DN-A)group(n=30)according to whether there was a large amount of proteinuria and the severity.Clinical data such as serum albumin(ALB),estimated glomerular filtration rate(eGFR)and albumin-to-creatinine ratio(ACR)were collected.Serum FGL1 level was detected by enzyme-linked immunosorbent assay(ELISA).Pearson linear correlation was used for correlation,the diagnostic value was analyzed by ROC curve.Results Compared with the control group,the levels of ACR,FGL1 in patients with DM group increased,the levels of eGFR decreased,and the differences were statistically significant(t=5.686,4.336,-4.683,all P<0.05).Compated with the DM group,the levels of ACR,FGL1 in patients with DN-E group was increased,and the level of eGFR was decreased,and the differences were statistically significant(t=5.275,3.454,-4.969,all P<0.05).Compared with the DN-E group,the levels of ACR,FGL1 in the DN-A group were increased,the levels of eGFR were decreased,and the differences were statistically significant(t=7.881,7.051,-5.596,all P<0.05).Serum FGL1 level was negatively correlated with ALB and eGFR(r=-0.638,-0.547,all P<0.05),and positively correlated with ACR(r=0.691,P<0.05).The AUC(95%CI),specificity and sensitivity of serum FGL1 level in the diagnosis of DN were 0.947(0.908～0.987),100%and 82.4%,respectively.Conclusion The level of serum FGL1 in DN and DM patients is high,and the level of serum FGL1 is closely related to the common metabolic indexes such as ALB,eGFR and ACR in the diagnosis of DN,which may have certain clinical diagnostic value.

Objective To explore the correlation between the expression level of serum fibrinogen-like protein 1(FGL1)and the indexes of metabolism and renal function in patients with diabetic nephropathy(DN)and diabetes mellitus(DM),and provide reference for clinical diagnosis and treatment.Methods From January 2017 to April 2023,30 patients with DM and treated in Jiangsu Province Hospital of Chinese Medicine were selected as the DM group,68 patients with DN were selected as the DN group,and 36 healthy subjects were selected as the control group.The DN group was further divided into the early DN(DN-E)group(n=38)and the late DN(DN-A)group(n=30)according to whether there was a large amount of proteinuria and the severity.Clinical data such as serum albumin(ALB),estimated glomerular filtration rate(eGFR)and albumin-to-creatinine ratio(ACR)were collected.Serum FGL1 level was detected by enzyme-linked immunosorbent assay(ELISA).Pearson linear correlation was used for correlation,the diagnostic value was analyzed by ROC curve.Results Compared with the control group,the levels of ACR,FGL1 in patients with DM group increased,the levels of eGFR decreased,and the differences were statistically significant(t=5.686,4.336,-4.683,all P<0.05).Compated with the DM group,the levels of ACR,FGL1 in patients with DN-E group was increased,and the level of eGFR was decreased,and the differences were statistically significant(t=5.275,3.454,-4.969,all P<0.05).Compared with the DN-E group,the levels of ACR,FGL1 in the DN-A group were increased,the levels of eGFR were decreased,and the differences were statistically significant(t=7.881,7.051,-5.596,all P<0.05).Serum FGL1 level was negatively correlated with ALB and eGFR(r=-0.638,-0.547,all P<0.05),and positively correlated with ACR(r=0.691,P<0.05).The AUC(95%CI),specificity and sensitivity of serum FGL1 level in the diagnosis of DN were 0.947(0.908～0.987),100%and 82.4%,respectively.Conclusion The level of serum FGL1 in DN and DM patients is high,and the level of serum FGL1 is closely related to the common metabolic indexes such as ALB,eGFR and ACR in the diagnosis of DN,which may have certain clinical diagnostic value.

Objective To examine the expression pattern of apoptosis signal-regulating kinase 1(ASK1)in the intestinal tissues of patients with Crohn's disease(CD),and analyze its mechanistic impact on intestinal epithelial barrier function and inflammatory responses.Methods Ileal tissue samples from Crohn's disease patients and healthy controls were collected.ASK1 protein level was assessed by immunohistochemistry,and its relationship with the Crohn's disease activity index(CDAI)was analyzed.A mouse model of acute colitis was constructed using TNBS,and subjected to qRT-PCR,Western blotting and immunohistochemistry for ASK1 expression,and the association between ASK1 expression and the disease activity index was examined.Lentivirus transfection was employed to create stable Caco-2 cell lines with altered ASK1 expression,and the intestinal barrier integrity and inflammation were assessed by measuring transepithelial electrical resistance(TEER),FITC-dextran leakage,and IL-6,IL-1β levels.Furthermore,the effects of ASK1 expression on Krüppel-like factor 4(KLF4)levels was examined using qRT-PCR and Western blotting.Results ASK1 was highly expressed in the ileum of CD patients and positively correlated with CDAI.In a TNBS-induced mouse model of acute colitis,ASK1 expression was up-regulated and positively correlated with DAI.Inflammation-induced ASK1 overexpression weakened the Caco-2 cell intestinal barrier,whereas ASK1 knockdown strengthened it.Moreover,ASK1 had the capability to enhance the expression of inflammatory factors.Additionally,ASK1 knockdown increased KLF4 expression,while overexpression decreased it,indicating a negative correlation between ASK1 and KLF4.Conclusion ASK1 expression is notably higher in CD and positively correlates with disease activity.ASK1 can influence intestinal barrier integrity and inflammatory factor expression,possibly through its impact on KLF4 expression.

Biomimetic wet-adhesive hydrogels mimic the adhesive properties of biological organisms to achieve strong bonding in moist environments.Compared to conventional medical adhesives,these materials are characterized by enhanced biocompatibility,robust ad-hesion,and adjustable physicochemical properties.Although biomimetic wet-adhesive hydrogels have been applied in oral mucosal drug delivery,intraoral wound management,and implant surgery,a systematic review is currently lacking.This article aims to summarize the wet-adhesion mechanisms of bio-inspired materials and their applications in various scenarios and to provide insights and methodolo-gies for the design of novel intraoral dressings.

Ultrasonography（US）is the first-line imaging modality recommended by domestic and international guidelines for liver tumor screening，owing to its non-invasive nature，real-time dynamic imaging capability，cost-effectiveness，and operational convenience. In recent years，the integration of artificial intelligence（AI）and medical imaging has emerged as a major research focus. By leveraging large-scale data training，AI models can automatically recognize and analyze input imaging data and generate predictive outcomes. Notably，AI-based ultrasound imaging technology has achieved breakthrough advancements in the diagnosis and treatment of liver tumors. These innovations significantly improve diagnostic accuracy，optimize treatment strategies，predict disease progression and prognosis，and monitor therapeutic response. This article provides a comprehensive review of the latest applications and research progress of AI in ultrasound-based diagnosis and treatment of liver tumors.

Since the introduction of monoamine oxidase and monoamine neurotransmitter reuptake inhibitors for the treatment of major depression in the 1950s, their strengths and limitations have been fully and accurately determined. Therefore, the development of novel drugs for the treatment of depression has become a priority for researchers who aim to address treatment resistance and improve patient outcomes. Panax ginseng C. A. Mey (P. ginseng, Ren Shen) is a Chinese medicine used to treat neurological and psychiatric disorders. Numerous studies have shown that ginsenosides, the primary active constituents of P. ginseng, exert a wide range of effects on the central nervous system. Recent studies have demonstrated that ginsenosides possess significant antidepressant properties in animal models. Ginsenosides, such as Rb1 and Rg1, are steroidal molecules, and steroid derivatives have been successfully used in anesthesia, epilepsy, and more recently, postpartum depression treatment. Based on these findings, ginsenosides are promising candidates for the treatment of depression. This raises the following question: What are the prospects of using ginsenosides to treat depression? To gain a clearer understanding, this review provides a comprehensive analysis of recent research on the antidepressant potential of ginsenosides, along with insights and suggestions for future development in this field.