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.

A growing body of research points out that gut microbiota plays a key role in tumor immunotherapy. By optimizing the composition of intestinal microbiota, it is possible to effectively improve immunotherapy resistance and enhance its therapeutic effect. This article comprehensively analyzes the mechanism of intestinal microbiota influencing tumor immunotherapy resistance, expounds the current strategies for targeted regulation of intestinal microbiota, such as traditional Chinese medicine and plant components, fecal microbiota transplantation, probiotics, prebiotics and dietary therapy, and explores the potential mechanisms of these strategies to improve patients' resistance to tumor immunotherapy. At the same time, the article also briefly discusses the prospects and challenges of targeting intestinal microbiota to improve tumor immunotherapy resistance, which provides a reference for related research to help the strategy research of reversing tumor immunotherapy resistance.

Objective To explore the value of serum D-dimer(D-D),fibrinogen(FIB),platelet(PLT),C-reactive pro-tein(CRP)and tissue plasminogen activator inhibitor(PAI)-1 levels in predicting lower extremity deep vein thrombosis(DVT)after hip joint surgery in the elderly.Methods A retrospective analysis was performed on 165 elderly patients with hip joint surgery admitted from February 2020 to May 2022,including 89 males and 76 females,aged from 60 to 75 years old with an average of(66.43±5.48)years,and there were 102 cases of femoral neck fracture and 63 cases of femoral head necrosis.Serum levels of D-D,FIB,PLT,CRP and PAI-1 tests were performed in all patients within 24 hours after admission,and the patients were divided into DVT group and non-DVT group according to whether they developed DVT.Results The levels of D-D,FIB,PLT,CRP,and PAI-1 in the DVT group were higher than those in the non-DVT group(P＜0.001).Spearman analysis showed that DVT was positively correlated with PLT,CRP,D-D,FIB,and PAI-1 levels(r=0.382,0.213,0.410,0.310,0.353,all P＜0.001).The results of binary Logistic regression analysis showed that D-D and PLT were independent factors affecting the occurrence of DVT(OR=0.038,0.960,P=0.032,0.011).The area under curve(AUC)of D-D,FIB,PLT,CRP,PAI-1,and the five combined predictions for DVT were 0.843,0.692,0.871,0.780,0.819,and 0.960,respectively.The AUC of the five combined predictions was higher than that of the single prediction(P＜0.05).Conclusion D-D,FIB,PLT,CRP and PAI-1 are effective in predicting DVT after hip surgery in the elderly,and the combined prediction of the five factors has higher effi-cacy.

Objective:To analyze the distribution characteristics of prognostic factors affecting recurrence in peripheral T-cell lymphoma(PTCL)patients with different levels of histone deacetylase(HDAC)based on latent class analysis.Methods:112 PTCL patients who were treated in our hospital from September 2012 to September 2019 were selected and divided into recurrence group and non-recurrence group.The clinical data of the two groups of patients were compared.Multivariate logistic regression was used to analyze the risk factors for recurrence.Latent class analysis was used to compare the distribution characteristics of prognostic factors affecting recurrence between the high-risk group and the low-risk group.Results:There were 87 patients(77.68％)in recurrence group and 25 patients(22.32％)in non-recurrence group.The result of multivariate logistic regression showed that ECOG score ≥2,Ann Arbor stage Ⅲ-Ⅳ,IPI score＞2,bone marrow involvement,elevated serum β2-microglobulin(β2-MG),short-term efficacy not reaching complete remission(CR)or partial remission(PR),and the high expression of HDAC were all independent risk factors for recurrence in patients with PTCL(P＜0.05).The recurrence rate of patients with high HDAC levels was significantly higher than that of patiens with low HDAC levels(P＜0.05).The results of cluster analysis showed that the risk of recurrence was obviously clustered,and the patients could be divided into high recurrence risk group(HDAC＞5 points)and low recurrence risk group(HDAC≤5 points).The results of latent class analysis showed that patients with multiple risk factors account for a higher proportion in the high recurrence risk group,compared with the low recurrence risk group(P＜0.05).Conclusion:There are differences in recurrence rates among PTCL patients with different HDAC levels and in distribution characteristics of risk factors between high recurrence risk and low recurrence risk groups.

Objective To design a nuclear emergency medical rescue simulation platform to solve the problmes in nuclear accident scenario simulation,fidelity of on-site medical treatment training and training organization.Methords A nuclear emergency medical rescue simulation platform was developed with the technologies of virtual reality(VR),augmented reality(AR),spatial positioning,full-body motion capture,gesture simulation and three-dimensional modeling,Unity3D platform,3D Studio Max software and Visual Studio,which mainly consisted of a hardware system and a software system.The hardware system was composed of the AR training equipment and VR training equipment,of which,the AR training equipment included a surface contamination detector simulator,an individual decontaminator simulator and AR glasses,and the VR training equipment was made up of VR glasses,large-space localization equipment,backpack computers,data gloves and full-body motion capture equipment.The software system involved in several modules for theoretical knowledge teaching and assessment,VR/AR practical teaching training and VR/AR evaluation.Results The system realized simulated collaborative training and evaluation for medical emergency response to nuclear accidents.Conclusion The platform constructed contributes to solving the practical problems in organizing and implementing education and training for nuclear emergency medical rescue,and the education and training are both enhanced greatly.[Chinese Medical Equipment Journal,2024,45(9):50-56]

Objective To investigate the influence of different delivery techniques,treatment sites and dose-volume algorithms on the results of three-dimensional dosimetric verification for intensity-modulated radiation therapy(IMRT)plans and the importance of individualized quality assurance(QA)evaluation standard for radiotherapy plans.Methods Totally 350 tumor patients receiving radiotherapy at some hospital from January 2017 to February 2022 had their three-dimensional dosimetric verification results of IMRT plans selected retrospectively and underwent data collection with COMPASS system,and then were grouped in terms of delivery technique(fixed-beam IMRT and volumetric modulated arc therapy),treatment site(neck,chest and abdomen)and dose-volume algorithm(anisotropic analytical algorithm and collapsed cone convolution algorithm).All the groups were compared based on the 3％/2 mm criterion with regard to the Gamma pass rate of 10％prescription dose area(GP10％),Gamma pass rate(GP50％)and mean Gamma index(μGI5o％)of 50％prescription dose area,dose of 95％target volume(D95％)and its mean dose(Dmean),parotid gland mean dose(Dmean),dose of 1％spinal cord volume(D1％),dose of 1％brain stem volume(D1％)of head and neck radiotherapy plan,heart and lung mean dose(Dmean)and dose of 1％spinal cord volume(D1％)of chest radiotherapy plan and bladder,rectum and femur mean dose of abdomen radiotherapy plan(Dmean).SPSS 26.0 software was used for statistical analysis.Results For different delivery techniques,significant differences were found in all the QA results except GP50％of abdomen radiotherapy plan(P＜0.05).For different treatment sites,the differences were statistically significant between the QA results of head and neck radiotherapy plan and abdomen plan and between those of chest radio-therapy plan and abdomen radiotherapy plan(P＜0.05),while were not significant between the QA results of head and neck radiotherapy plan and chest radiotherapy plan(P＞0.05).For different dose-volume algorithms,the QA results had significant differences except D5％of abdomen radiotherapy planning target volume and Dmean and D1％of chest radiotherapy PTV(P＜0.05).Conclusion Dosimetric verification results vary depending on the delivery technique,treatment site and dose-volume algorithm.Statistical process control recommended by AAPM TG-218 report may be involved in to establish individualized QA standard for radiotherapy plans in case universal action limits are not appropriate.[Chinese Medical Equipment Journal,2024,45(11):54-59]

Objective:To investigate the correlation between serum albumin, urea nitrogen and Fazekas scores and cognitive function scores in patients with mild and medium ischemic stroke.Methods:Clinical data of 160 patients with acute ischemic stroke with the National Institutes of Health Stroke Scale (NIHSS)≤7 scores admitted to the Department of Neurology of the First Affiliated Hospital of Hainan Medical College from June 2021 to April 2023 were selected for a cross-sectional study. According to the Montreal Cognitive Assessment (MoCA) score, they were divided into normal cognitive group (28 cases) (MoCA≥26 scores), mild to moderate cognitive impairment group (74 cases) (MoCA 15-<26 scores), and severe cognitive impairment group (58 cases) (MoCA<15 scores). Demographic characteristics, serological indicators and imaging data of patients were collected, and the correlation between serum albumin, urea nitrogen and Fazekas scores and the total score of MoCA and the scores of each cognitive domain was analyzed. One-way ANOVA was used for comparison between the normal distribution and homogeneous variance data sets, LSD analysis was used for pairwise comparison, Kruskal-Wallis H test was used between the skew distribution or heterogeneous variance data sets. Bonferroni correction analysis was used for pairwise comparison. Chi-square test or Fisher exact probability method was used after the comparison between the count data sets. Spearman Spearman correlation analysis was performed on serum albumin, urea nitrogen and Fazekas scores with MoCA scores and cognitive domain scores. Multivariate ordered Logistic regression was used to analyze the independent risk factors of cognitive function in acute stage of mild and medium ischemic stroke patients. Results:The incidence of cognitive impairment in patients with acute mild and medium ischemic stroke was 82.50% (132/160). Comparison of age ((56.71±7.35), (60.32±10.20), (66.40±11.88) years old), sex (male/female: (23/5, 58/16, 33/25)), the proportion of education level above high school (25.0%(7/28), 16.2%(12/74), 6.9%(4/58)), hemoglobin ((149.26±14.91), (144.85±16.85), (137.63±17.22) g/L), albumin (39.5 (37.0, 41.2), 38.6(35.6, 40.8), 37.4 (34.5, 39.8) g/L), urea nitrogen (5.30 (4.00, 6.60), 4.81 (4.00, 6.32), 5.86 (4.55, 6.97) mmol/L), Hamilton Anxiety Scale (HAMA) score (5.0 (2.0, 10.0), 7.5 (5.0, 11.0), 10.0 (6.0, 14.3) scores),Hamilton Depression Scale (HAMA) score (5.5 (3.0, 12.5), 7.0 (4.0, 11.0), 9.5 (5.0, 14.0) scores), and Fazekas score (2.00 (1.25, 3.00), 2.00 (1.00, 4.00), 3.00 (2.00, 5.00) scores) among cognitive normal group, mild to moderate cognitive impairment group, and severe cognitive impairment group of patients, the difference were statistically significant (the statistical values were F=9.68, χ 2=9.29, χ 2=30.77, F=5.31, H=7.06, H=6.71, H=12.37, H=8.91, and H=10.96, respectively;the P values were <0.001, 0.010, <0.001, 0.006, 0.029, 0.035, 0.002, 0.012, and 0.004, respectively ). The total score of MoCA was negatively correlated with Fazekas score and serum urea nitrogen, but positively correlated with serum albumin ( r s values were -0.250, -0.168, and 0.212, respectively; P values were 0.001, 0.036, and 0.009, respectively). Serum albumin was positively correlated with scores in visual space and execution, naming, attention and orientation, serum urea nitrogen was negatively correlated with scores in language and orientation, and Fazekas score was negatively correlated with scores in visual space and execution, orientation, attention and language ( r s values were 0.291, 0.196, 0.191, 0.209, -0.205, -0.180, -0.248, -0.193, -0.188, and -0.183, respectively; P values were <0.001, 0.017, 0.020, 0.011, 0.012, 0.027, 0.002, 0.016, 0.020, and 0.023, respectively). Multiple Logistic regression analysis showed that low albumin ( OR=0.884, 95% CI: 0.813-0.963, P=0.005) and high urea nitrogen ( OR=1.195, 95% CI: 1.003-1.425, P=0.047) and high Fazekas scores ( OR=1.401, 95% CI: 1.132-1.733, P=0.002) were independent risk factors for cognitive function, while high education level was a protective factor ( OR=0.062, 95% CI: 0.019-0.202, P<0.001). Conclusion:The incidence of acute cognitive impairment is high in patients with mild and medium ischemic stroke. Higher education level is a protective factor for cognitive function. Low albumin, high urea nitrogen and high Fazekas score are independent risk factors for cognitive function.

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*

ObjectiveTo explore the mechanism of treadmill exercise against type 2 diabetes mellitus (T2DM) with non-alcoholic fatty liver disease (NAFLD) based on the regulator effects of exercise on ferroptosis. MethodsEight 8-week-old male m/m mice were used as control group (Con, n=8), and db/db mice of the matched age were randomly divided into T2DM model group (db/db, n=8), exercise group (db+Exe, n=8), p38 mitogen-activated protein kinase (MAPK) inhibitor group (db+SB203580, n=8) and exercise combined with p38 MAPK inhibitor group (db+Exe+SB203580, n=8). After one-week adaptive feeding, the mice in the db+Exe group and db+Exe+SB203580 group underwent moderate intensity treadmill exercise for 40 min/d, 5 d/week lasting 8 weeks. The db+SB203580 group and db+Exe+SB203580 group were treated with SB203580 (a specific inhibitor of p38 MAPK) with a dose of 5 mg/kg, 5 d/week for 8 weeks. And the exercise intervention was performed 2 h later after the intraperitoneal injection of SB203580. The body weight and fasting blood glucose of mice were measured regularly every week during the experiment. After 24 h of the last intervention, the mice were weighted, the liver tissues were taken, weighted and the liver index was calculated. The pathological changes of liver were determined by Oil Red O and hematoxylin-eosin (HE) staining. The levels of blood lipids, liver function, Fe²⁺ and oxidative stress markers of liver were measured by enzyme linked immunosorbent assay (ELISA). The related mRNA expression levels of lipogenesis and inflammation were evaluated by quantitative reverse transcriptase-mediated PCR (qRT-PCR). The related protein expression levels of lipogenesis and ferroptosis in liver were determined by immunohistochemical (IHC) staining and Western blot. ResultsThe body weight, fasting blood glucose, liver index, blood lipid and transaminase levels in the db/db group were significantly increased compared with the Con group. HE and Oil Red O staining showed severe lipid accumulation and ballooning change in the liver of db/db mice. Biochemical tests showed that Fe²⁺ and MDA level of liver constitution homogenate increased, while GSH level decreased significantly. The results of qRT-PCR showed that the mRNA levels of MCP-1, IL-6, SREBF1 and ACC1 in liver tissue of db/db mice were all significantly increased. Western blot results showed that the expression levels of SREBF1, ACC1 increased, ferroptosis relative proteins were significantly decreased. The 8 weeks of exercise significantly reduced the rise in body weight, blood glucose, liver index and blood lipid levels in db/db mice. Exercise intervention also alleviated hepatic steatosis and reduced the expression levels of Fe²⁺, MDA, MCP-1, IL-6, ACC1 and SREBF1, upregulated the expression levels of GSH, NRF2, HO-1, SLC7A11 and GPX4 in liver tissue of db/db mice. The intervention of exercise combined with SB203580 significantly down-regulated the mRNA expression levels of ACC1, MCP-1, IL-6, reduced the levels of Fe²⁺ and MDA, and up-regulated the level of GSH in db/db mice. Compared with the db+Exe group, the expression of Fe²⁺, MDA, MCP-1, and SREBF1 in the liver of the db+Exe+SB203580 group mice significantly increased, while the expression level of GSH and expression levels of ferroptosis relative proteins also significantly decreased. In addition, compared with db+SB203580 group, the iron accumulation and lipid peroxidation in the liver of db+Exe+SB203580 group were significantly improved. ConclusionThe8-week treadmill exercise can effectively alleviate liver injury and steatosis, and its mechanism may be related to the inhibition of hepatocyte ferroptosis through p38 MAPK signal.

Pancreatic cancer is one of the most lethal malignant tumors. While the existing surgical resection and chemotherapy have failed to improve the prognosis of patients well, it is urgent to find safer and more effective therapies to meet the clinical therapeutic needs. Antibody drug conjugate (ADC) is a class of targeted antitumor agents that combines monoclonal antibodies with small molecule cytotoxic drugs via chemical linkers. ADC provided wider therapeutic windows and enhanced pharmacokinetic/pharmacodynamic properties which depended on their characteristics of high selectivity, high efficacy and low toxicity. In recent years, the successful application of ADC in the treatment of multiple tumors has aroused a research upsurge in ADC for pancreatic cancer. This review summarizes the structure and mechanism of ADC and the current status of ADC for pancreatic cancer in order to provide some references for the future development of ADC for pancreatic cancer.