ObjectiveTo construct a large language model （LLM）-based intelligent pre-consultation system for traditional Chinese medicine （TCM） to improve efficacy of clinical practice. MethodsA TCM large language model was fine-tuned using DeepSpeed ZeRO-3 distributed training strategy based on YAYI 2-30B. A weighted undirected graph network was designed and an agent-based syndrome differentiation model was established based on relationship data extracted from TCM literature and clinical records. An agent collaboration framework was developed to integrate the TCM LLM with the syndrome differentiation model. Model performance was comprehensively evaluated by Loss function， BLEU-4， and ROUGE-L metrics， through which training convergence， text generation quality， and language understanding capability were assessed. Professional knowledge test sets were developed to evaluate system proficiency in TCM physician licensure content， TCM pharmacist licensure content， TCM symptom terminology recognition， and meridian identification. Clinical tests were conducted to compare the system with attending physicians in terms of diagnostic accuracy， consultation rounds， and consultation duration. ResultsAfter 100 000 iterations， the training loss value was gradually stabilized at about 0.7±0.08， indicating that the TCM-LLM has been trained and has good generalization ability. The TCM-LLM scored 0.38 in BLEU-4 and 0.62 in ROUGE-L， suggesting that its natural language processing ability meets the standard. We obtained 2715 symptom terms， 505 relationships between diseases and syndromes， 1011 relationships between diseases and main symptoms， and 1 303 600 relationships among different symptoms， and constructed the Agent of syndrome differentiation model. The accuracy rates in the simulated tests for TCM practitioners， licensed pharmacists of Chinese materia medica， recognition of TCM symptom terminology， and meridian recognition were 94.09%， 78.00%， 87.50%， and 68.80%， respectively. In clinical tests， the syndrome differentiation accuracy of the system reached 88.33%， with fewer consultation rounds and shorter consultation time compared to the attending physicians （P＜0.01）， suggesting that the system has a certain pre- consultation ability. ConclusionThe LLM-based intelligent TCM pre-diagnosis system could simulate diagnostic thinking of TCM physicians to a certain extent. After understanding the patients' natural language， it collects all the patient's symptom through guided questioning， thereby enhancing the diagnostic and treatment efficiency of physicians as well as the consultation experience of the patients.

Aim To explore the effect of astragalosideⅣ(AS-Ⅳ)on lipopolysaccharide(LPS)-induced po-larization and migration of RAW 264.7 macrophages and the underlying mechanism.Methods 1 mg·L-1 LPS was used to construct cell migration model.Scratch assay was utilized to determine cell migration rate.Immunofluorescence staining was utilized to de-tect the expression and location of F4/80,iNOS and Arg-1.CCK-8 assay was used to determine the viabili-ty of RAW 264.7 cells.Griess assay was used to measure NO content.Molecular docking was used to analyze the interaction between AS-Ⅳ and the core tar-gets such as cGAS and STING protein.Western blot was employed to detect the expression of iNOS,Arg-1,cGAS,STING,NF-κB p65 and p-NF-κB p65 protein.Results AS-Ⅳ significantly inhibited the migration and M1 polarization of RAW 264.7 cells induced by LPS.Moreover,AS-Ⅳ could interact with cGAS and STING protein,especially cGAS.Further Western blot assay showed that AS-Ⅳ significantly downregulated the expression of iNOS,cGAS,STING and p-NF-κB p65 protein.Conclusions AS-Ⅳ could promote mac-rophage M1 to M2 polarization,thereby inhibited mac-rophage migration through restraining the cGAS/STING/NF-κB signaling pathway,which provides a new therapeutic target for AS-Ⅳ to improve the early inflammatory response of AS.

Medical engineering in TCM medical institutions in Guizhou Province was described in terms of its development status and problems in functional positioning,professional title appraisal,unbalanced staff composition and discipline foun-dation.Some suggestions were put forward including determining the functional positioning and management mode of medical engineering institutions,promoting the professional title system for medical engineering staffs,strengthening medical engineering team and enhancing medical engineering discipline.References were provided for the development of medical engineering in TCM medical institutions in Guizhou Province.[Chinese Medical Equipment Journal,2025,46(5):84-90]

Objective To explore the effect of endoscopic ultrasonography(EUS)with different probes on the measurement of gastrointestinal submucosal tumor(SMT)diameter.Methods The clinical data of 356 patients(with a total of 372 lesions)initially diagnosed as SMT by EUS at the Second Affiliated Hospital of Army Medical University from January 2023 to June 2024 were analyzed retrospectively.The basic characteristics of the origin layers and pathological distribution of SMT were analyzed.The differences between the EUS-measured diameters and the actual diameters of SMT at different diameter ranges of lesions were compared.Taking the postoperative pathological diagnosis results as the gold standard,the effects of different clinicopathological features and probe types on the relative error in EUS-measured diameters of SMT were analyzed.Results Among the 372 gastrointestinal SMT lesions,lesions were more frequent in female patients,gastric lesions was the most common,and leiomyoma was the predominant pathological type.The accuracy of EUS in diagnosing SMT was 94.4%.A statistically significant difference was observed between the EUS-measured diameters and the actual diameters of SMT(P<0.05).There were significantly differences in various ranges of lesions between the EUS-measured diameters and the actual diameters(P<0.05).The gender,age,lesion location,and pathological type had no significant effect on the relative error of EUS-measured diameters(P>0.05);while probe types had a significant effect on the relative error of EUS-measured diameters(P<0.05).For lesions with an actual diameter of 2.0 to 3.9 cm,the relative error of SMT diameters measured by small-probe EUS was significantly greater than that by large-probe EUS(P<0.05).Conclusion Large-probe EUS exhibits a smaller relative error in measuring the diameter of SMT with a diameter of≥2.0 cm.Therefore,large-probe EUS is recommended for the examination of SMT with an estimated diameter exceeding 2.0 cm.

Objective To explore the effect of endoscopic ultrasonography(EUS)with different probes on the measurement of gastrointestinal submucosal tumor(SMT)diameter.Methods The clinical data of 356 patients(with a total of 372 lesions)initially diagnosed as SMT by EUS at the Second Affiliated Hospital of Army Medical University from January 2023 to June 2024 were analyzed retrospectively.The basic characteristics of the origin layers and pathological distribution of SMT were analyzed.The differences between the EUS-measured diameters and the actual diameters of SMT at different diameter ranges of lesions were compared.Taking the postoperative pathological diagnosis results as the gold standard,the effects of different clinicopathological features and probe types on the relative error in EUS-measured diameters of SMT were analyzed.Results Among the 372 gastrointestinal SMT lesions,lesions were more frequent in female patients,gastric lesions was the most common,and leiomyoma was the predominant pathological type.The accuracy of EUS in diagnosing SMT was 94.4%.A statistically significant difference was observed between the EUS-measured diameters and the actual diameters of SMT(P<0.05).There were significantly differences in various ranges of lesions between the EUS-measured diameters and the actual diameters(P<0.05).The gender,age,lesion location,and pathological type had no significant effect on the relative error of EUS-measured diameters(P>0.05);while probe types had a significant effect on the relative error of EUS-measured diameters(P<0.05).For lesions with an actual diameter of 2.0 to 3.9 cm,the relative error of SMT diameters measured by small-probe EUS was significantly greater than that by large-probe EUS(P<0.05).Conclusion Large-probe EUS exhibits a smaller relative error in measuring the diameter of SMT with a diameter of≥2.0 cm.Therefore,large-probe EUS is recommended for the examination of SMT with an estimated diameter exceeding 2.0 cm.

Medical engineering in TCM medical institutions in Guizhou Province was described in terms of its development status and problems in functional positioning,professional title appraisal,unbalanced staff composition and discipline foun-dation.Some suggestions were put forward including determining the functional positioning and management mode of medical engineering institutions,promoting the professional title system for medical engineering staffs,strengthening medical engineering team and enhancing medical engineering discipline.References were provided for the development of medical engineering in TCM medical institutions in Guizhou Province.[Chinese Medical Equipment Journal,2025,46(5):84-90]

Aim To explore the effect of astragalosideⅣ(AS-Ⅳ)on lipopolysaccharide(LPS)-induced po-larization and migration of RAW 264.7 macrophages and the underlying mechanism.Methods 1 mg·L-1 LPS was used to construct cell migration model.Scratch assay was utilized to determine cell migration rate.Immunofluorescence staining was utilized to de-tect the expression and location of F4/80,iNOS and Arg-1.CCK-8 assay was used to determine the viabili-ty of RAW 264.7 cells.Griess assay was used to measure NO content.Molecular docking was used to analyze the interaction between AS-Ⅳ and the core tar-gets such as cGAS and STING protein.Western blot was employed to detect the expression of iNOS,Arg-1,cGAS,STING,NF-κB p65 and p-NF-κB p65 protein.Results AS-Ⅳ significantly inhibited the migration and M1 polarization of RAW 264.7 cells induced by LPS.Moreover,AS-Ⅳ could interact with cGAS and STING protein,especially cGAS.Further Western blot assay showed that AS-Ⅳ significantly downregulated the expression of iNOS,cGAS,STING and p-NF-κB p65 protein.Conclusions AS-Ⅳ could promote mac-rophage M1 to M2 polarization,thereby inhibited mac-rophage migration through restraining the cGAS/STING/NF-κB signaling pathway,which provides a new therapeutic target for AS-Ⅳ to improve the early inflammatory response of AS.

AIM To investigate the effects of Ginkgo biloba extract on hearing function,cochlear morphology and autophagy-related protein expression in a rat model of presbycusis.METHODS Forty-five rats were randomly divided into the control group,the model group and the low,medium and high dose G.biloba extract groups(10,20 and 30 mg/kg),with 9 rats in each group.The rat model of presbycusis was established by intraperitoneal injection of 500 mg/kg D-galactose(D-gal).Eight weeks after the corresponding administration,the rats had their changes of hearing threshold detected by the auditory brainstem evoked potential(ABR);their morphological changes of cochlear hair cells,stria vascularis(SV)and spiral ganglion cells observed by HE staining;their number of hair cells inside and outside the cochlea detected by immunofluorescence staining;their ultrastructure changes of cochlear hair cells observed by transmission electron microscopy;and their expression of autophagy-related proteins in cochlea tissue detected by Western blot.RESULTS Compared with the control group,the model group displayed increased ABR threshold(P<0.01);more severely damaged inner and outer hair cells,spiral ganglion cells and SV,decreased SV thickness and numbers of spiral ganglion cells,inner and outer hair cells and autophagosomes(P<0.01);decreased protein expressions of Beclin1 and LC3 Ⅱ and ratio of LC3 Ⅱ/LC3 Ⅰ in cochlear tissue(P<0.01),and higher P62 protein expression(P<0.01).Compared with the model group,the medium and high dose G.biloba extract groups shared decreased ABR thresholds(P<0.01);improved morphology of inner and outer hair cells and SV in the cochlea,normalized,morphology of spiral ganglion cells,and increased SV thickness and the numbers of spiral ganglion cells,inner and outer hair cells and autophagosomes(P<0.05,P<0.01);increased protein expressions of Beclin1 and LC3 Ⅱ and the ratio of LC3 Ⅱ/LC3 Ⅰ in the cochlea(P<0.01),and decreased P62 protein expression(P<0.01).CONCLUSION The protective effects G.biloba extract on hearing function and cochlear cells in the rat model of presbycusis may be associated with the up-regulated expression of Beclin1 and LC3 Ⅱ proteins and down-regulated P62 protein expression in cochlear tissues.

OBJECTIVE: To study the in vitro and in vivo antitumor effects of the polysaccharide of Alocasia cucullata (PAC) and the underlying mechanism. METHODS: B16F10 and 4T1 cells were cultured with PAC of 40 µg/mL, and PAC was withdrawn after 40 days of administration. The cell viability was detected by cell counting kit-8. The expression of Bcl-2 and Caspase-3 proteins were detected by Western blot and the expressions of ERK1/2 mRNA were detected by quantitative real-time polymerase chain reaction (qRT-PCR). A mouse melanoma model was established to study the effect of PAC during long-time administration. Mice were divided into 3 treatment groups: control group treated with saline water, positive control group (LNT group) treated with lentinan at 100 mg/(kg·d), and PAC group treated with PAC at 120 mg/(kg·d). The pathological changes of tumor tissues were observed by hematoxylin-eosin staining. The apoptosis of tumor tissues was detected by TUNEL staining. Bcl-2 and Caspase-3 protein expressions were detected by immunohistochemistry, and the expressions of ERK1/2, JNK1 and p38 mRNA were detected by qRT-PCR. RESULTS: In vitro, no strong inhibitory effects of PAC were found in various tumor cells after 48 or 72 h of administration. Interestingly however, after 40 days of cultivation under PAC, an inhibitory effect on B16F10 cells was found. Correspondingly, the long-time administration of PAC led to downregulation of Bcl-2 protein (P<0.05), up-regulation of Caspase-3 protein (P<0.05) and ERK1 mRNA (P<0.05) in B16F10 cells. The above results were verified by in vivo experiments. In addition, viability of B16F10 cells under long-time administration culture in vitro decreased after drug withdrawal, and similar results were also observed in 4T1 cells. CONCLUSIONS Long-time administration of PAC can significantly inhibit viability and promote apoptosis of tumor cells, and had obvious antitumor effect in tumor-bearing mice.
Mice ; Animals ; Alocasia/metabolism* ; MAP Kinase Signaling System ; Caspase 3/metabolism* ; Apoptosis ; RNA, Messenger/metabolism*

Objective To construct an artificial intelligence assisted diagnosis model based on deep learning for dynamically recognizing gastric lesions and their locations under gastroscopy in real time,and to evaluate its ability to detect and recognize gastric lesions and their locations.Methods The gastroscopy videos of 104 patients in our hospital was retrospectively analyzed,and the video frames were manually annotated.The annotated picture frames of lesion category were divided into the training set and the validation set according to the ratio of 8∶2,and the annotated picture frames of location category were divided into the training set and the validation set according to the patient sources at the ratio of 8∶2.These sets were utilized for training and validating the respective models.YoloV4 model was used for the training of lesion recognition,and ResNet152 model was used for the training of location recognition.The accuracy,sensitivity,specificity,positive predictive value,negative predictive value and location recognition accuracy of the auxiliary diagnostic model were evaluated.Results A total of 68 351 image frames were annotated,with 54 872 frames used as the training set,including 41 692 frames for lesion categories and 13 180 frames for location categories.The validation set consisted of 13 479 frames,comprising 10 422 frames for lesion categories and 3 057 frames for location categories.The lesion recognition model achieved an overall accuracy of 98.8%,with a sensitivity of 96.6%,specificity of 99.3%,positive predictive value of 96.3%,and negative predictive value of 99.3% in validation set.Meanwhile,the location recognition model demonstrated an top-5 accuracy of 87.1% .Conclusion The artificial intelligence assisted diagnosis model based on deep learning for real-time dynamic recognition of gastric lesions and their locations under gastroscopy has good ability in lesion detection and location recognition,and has great clinical application prospects.