Objective To explore the application value of artificial intelligence in medical research assistance, and analyze the key paths to achieve precise execution of model instructions, improvement of model interpretation completeness, and control of hallucinations. Methods Taking esophageal cancer research as the scenario, five types of literature including research articles, case reports, reviews, editorials, and guidelines were selected for model interpretation tests. The model performance was systematically evaluated from five dimensions: recognition accuracy, format accuracy, instruction execution accuracy, content reliability rate, and content completeness index. The performance differences of Ruibin Agent, GPT-4o, Claude 3.7 Sonnet, DeepSeek V3, and DouBao-pro models in medical literature interpretation tasks were compared. Results A total of 15 studies were included, with 3 studies of each type. The five models collectively conducted 1 875 tests. Due to the poor recognition accuracy of the editorial type, the overall recognition accuracy of Ruibin Agent was significantly lower than other models (92.0% vs. 100.0%, P＜0.001). In terms of format accuracy, Ruibin Agent was significantly better than Claude 3.7 Sonnet (98.7% vs. 92.0%, P=0.002) and GPT-4o (98.7% vs. 78.9%, P＜0.001). In terms of instruction execution accuracy, Ruibin Agent was better than GPT-4o (97.3% vs. 80.0%, P＜0.001). In terms of content reliability rate, Ruibin Agent was significantly lower than Claude 3.7 Sonnet (84.0% vs. 92.0%, P=0.010) and DeepSeek V3 (84.0% vs. 94.7%, P＜0.001). In terms of content completeness index, the median scores of Ruibin Agent, GPT-4o, Claude 3.7 Sonnet, DeepSeek V3, and DouBao-pro were 0.71, 0.60, 0.85, 0.74, and 0.77, respectively. Conclusion Ruibin Agent has significant advantages in terms of formatted interpretation of medical literature and instruction execution accuracy. In the future, it is necessary to focus on optimizing the recognition ability of editorial types, strengthening the coverage ability of core elements of various types of literature to improve interpretation completeness, and improving content reliability through optimizing the confidence mechanism to ensure the rigor of medical literature interpretation.

OBJECTIVE: To investigate whether auraptene (AUR) exerts a protective effect on acute diquat (DQ)-induced liver injury in mice and explore its underlying mechanisms. METHODS: Forty SPF-grade healthy male C57BL/6 mice were randomly divided into normal control group (Control group), DQ poisoning model group (DQ group), AUR treatment group (DQ+AUR group), and AUR control group (AUR group), with 10 mice in each group. The DQ poisoning model was established via a single intraperitoneal injection of 40 mg/kg DQ aqueous solution (0.5 mL); Control group and AUR group received an equal volume of pure water intraperitoneally. Four hours post-modeling, DQ+AUR group and AUR group were administered 0.5 mg/kg AUR aqueous solution (0.2 mL) by gavage once daily for 7 consecutive days, while Control group and DQ group received pure water. Blood and liver tissues were collected after anesthesia on day 7. Liver ultrastructure was observed by transmission electron microscopy. Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels were measured via enzyme-linked immunosorbent assay (ELISA). Hepatic glutathione (GSH), superoxide dismutase (SOD), and malondialdehyde (MDA) levels were detected using WST-1, thiobarbituric acid (TBA), and enzymatic reaction methods, respectively. Protein expression of nuclear factor-erythroid 2-related factor 2 (Nrf2), heme oxygenase-1 (HO-1), Kelch-like ECH-associated protein 1 (Keap1), and activated caspase-9 in liver tissues was analyzed by Western blotting. RESULTS: Transmission electron microscopy revealed that mitochondria in the Control group exhibited mild swelling, uneven distribution of matrix, and a small number of cristae fractures. In the AUR group, mitochondria showed mild swelling, with no obvious disruption of cristae structure. In the DQ group, mitochondria demonstrated marked swelling and increased volume, matrix dissolution, loss and fragmentation of cristae, and extensive vacuolization. In contrast, the DQ+AUR group showed significantly reduced mitochondrial swelling, volume increase, matrix dissolution, cristae loss and fragmentation, and vacuolization compared to the DQ group. Compared with the DQ group, the DQ+AUR group exhibited significantly lower serum AST levels (U/L: 173.45±23.60 vs. 255.33±41.51), ALT levels (U/L: 51.77±21.63 vs. 100.70±32.35), and hepatic MDA levels (μmol/g: 12.40±2.76 vs. 19.74±4.10), along with higher hepatic GSH levels (mmol/g: 37.65±14.95 vs. 20.58±8.52) and SOD levels (kU/g: 124.10±33.77 vs. 82.81±22.00), the differences were statistically significant (all P < 0.05). Western blotting showed upregulated Nrf2 expression (Nrf2/β-actin: 0.87±0.37 vs. 0.53±0.22) and HO-1 expression (HO-1/β-actin: 1.06±0.22 vs. 0.49±0.08), and downregulated Keap1 expression (Keap1/β-actin: 0.82±0.12 vs. 1.52±0.76) and activated caspase-9 expression (activated caspase-9/β-actin: 1.16±0.28 vs. 1.71±0.30) in the DQ+AUR group compared to the DQ group (all P < 0.05). CONCLUSION AUR attenuates DQ-induced acute liver injury in mice by activating the Keap1/Nrf2 signaling pathway.
Animals ; Male ; Mice ; Mice, Inbred C57BL ; Liver/pathology* ; Chemical and Drug Induced Liver Injury/drug therapy* ; Diquat/poisoning* ; NF-E2-Related Factor 2/metabolism* ; Oxidative Stress ; Apoptosis ; Coumarins

Evading host immunity killing is a critical step for virus survival. Inhibiting viral immune escape is crucial for the treatment of viral diseases. Serine/threonine kinase 39 (STK39) was reported to play an essential role in ion homeostasis. However, its potential role and mechanism in viral infection remain unknown. In this study, we found that viral infection promoted STK39 expression. Consequently, overexpressed STK39 inhibited the phosphorylation of interferon regulatory factor 3 (IRF3) and the production of type I interferon, which led to viral replication and immune escape. Genetic ablation or pharmacological inhibition of STK39 significantly protected mice from viral infection. Mechanistically, mass spectrometry and immunoprecipitation assays identified that STK39 interacted with PPP2R1A (a scaffold subunit of protein phosphatase 2A (PP2A)) in a kinase activity-dependent manner. This interaction inhibited DDB1 and CUL4 associated factor 1 (DCAF1)-mediated PPP2R1A degradation, maintained the stabilization and phosphatase activity of PP2A, which, in turn, suppressed the phosphorylation of IRF3, decreased the production of type I interferon, and then strengthened viral replication. Thus, our study provides a novel theoretical basis for viral immune escape, and STK39 may be a potential therapeutic target for viral infectious diseases.

OBJECTIVE To explore the potential mechanisms and bioactive compounds of licorice against COVID-19 based on a multidimensional interaction of"component-disease-symptom-target".METHODS Firstly,candidate target sets for licorice compo-nents were obtained from the ETCM2.0 database based on the Encyclopedia of Traditional Chinese Medicine,and the disease symp-toms and associated target sets for COVID-19 were derived from the GeneCards and HPO databases.And then a protein interaction network was constructed using the String database(version 12.0).By calculating topological eigenvalues and functional enrichment analysis,the potential mechanisms of action were explored.Combined with Schr?dinger molecular docking virtual screening,candidate pharmacodynamic substances were identified.Fluorescence resonance energy transfer was used for experimental verification.RESULTS Licorice might improve symptoms of COVID-19(fever,chest pain and so on)by regulating the imbalance of"immune-inflammation"network and signal transduction abnormalities during the development and progress of COVID-19,such as coronavirus disease-COVID-19,Toll-like receptor signaling pathway and NOD-like receptor signaling pathway.The components,such as Isos-chaftoside and Hesperidin,were identified to possess high binding affinity with the critical enzymes for viral replication.Isoschaftoside,Hesperidin,Isoliquiritin apioside and Vicenin-2 exhibited varying degrees of inhibition on the enzyme of 3CLpro at different concentra-tions while excluding the interference of the compounds' fluorescence.CONCLUSION Through experimental verification using a multidimensional interaction network of"component-disease-symptom-target",the key pharmacologically active substances of licorice in the fight against COVID-19 are preliminarily identified,and their mechanism of action through overall regulation is elucidated.

Objective:To explore the predictive value of systemic immune inflammation index combined with stress response index for postoperative urinary tract infection in patients with complicated kidney stones.Methods:From June 2021 to June 2023,97 patients who underwent treatment for urinary tract infection after complex kidney stone operation in Cangzhou Hospital of Integrated Traditional and Western Medicine were selected as infected group,and 87 patients who did not develop urinary tract infection after complex kidney stone operation were selected as uninfected group.Systemic immunoinflammatory index(SII),neutrophils,lympho-cyte,platelet,malondialdehyde(MDA),superoxide dismutase(SOD),catalase(CAT)levels were detected.Multivariate Logistic regression analysis was performed to analyze risk factors of postoperative urinary tract infection,and ROC curve was drawn to analyze the predictive value of SII,MDA,SOD and CAT alone and combined detection for postoperative urinary tract infection in patients with complex kidney stones.Results:Compared with uninfected group,levels of SII,neutrophils and MDA were increased in infected group,while levels of SOD and CAT were decreased(P<0.05).Levels of lymphocytes and platelets were decreased,the difference was not statistically significant(P>0.05).Multivariate Logistic regression analysis showed that presence of urinary tract history,opera-tion time≥100 min,urinary catheter retention time≥7 d,presence of preoperative urinary tract infection,stone load≥1 000 mm2,combined renal dysfunction,and preoperative blood glucose≥6.15 mmol/L were main risk factors for postoperative urinary tract infection in patients with complex kidney stones.ROC curve showed that combined detection was significantly more effective than single detec-tion of SII,MDA,SOD and CAT in the diagnosis of postoperative urinary tract infection in patients with precomplex kidney stones.Conclusion:Patients with urinary tract infection after complicated kidney stones have increased SII and MDA,decreased SOD and CAT levels,and the abnormal increased or decreased expression level are the predictors of risk of urinary tract infection after compli-cated kidney stones,which may be related with the diagnosis,development and prognosis of the disease.

Microfibrino-associated protein (MFAP) 2 is an extracellular matrix glycoprotein and a member of the MFAP family. It participates in the assembly of extracellular elastic microfibers.Upregulation of MFAP2 can promote the occurrence and development of various tumors and regulate multiple cancer-related signaling pathways and is related to their prognosis, making it a potential target for tumor treatment. This article summarizes the research progress on the pathogenesis, targeted therapy and prognosis of MFAP2 in malignant tumors.

Objective To address the challenges of accurately capturing critical features in small-sample diabetic retinopathy(DR)recognition models in real-world applications,and the overly smooth distribution of true and false feature coefficients,we propose an enhanced small-sample DR recognition method based on an improved capsule network.Methods Firstly,the method enhances image feature representation by removing redundant boundary information and employing discrete wavelet transform based on the Haar wavelet function,thereby high-lighting critical pathological features.Secondly,the convolutional layer of the capsule network is optimized through a multi-branch architecture to extract multi-scale features from retinal images,while incorporating a convolutional block attention module that is subsequently fed into the capsule layer.Finally,the sigmoid function replaces the soft-max function in dynamic routing,thereby improving the model's robustness.Result The enhanced neural network model achieved an accuracy of 98.62%on the Kaggle public dataset following a rigorous selection and preprocessing procedure.Conclusion The enhanced capsule network demonstrated superior precision in identifying diabetic reti-nopathy within small sample sizes compared to other state-of-the-art algorithms currently available.

Objective:To develop a scientific and practical questionnaire for general practice residents, and to conduct multidimensional and comprehensive evaluation of the quality of general practice teaching clinics.Methods:A preliminary draft of the questionnaire items was formulated based on a literature review and in-depth interviews. The Delphi method was employed to conduct two rounds of consultation with 14 experts. Following revisions, a convenience sampling method was used to invite general practice residents from three standardized residency training bases to test the reliability and validity of the questionnaire.Results:The questionnaire consisted of 23 items, covering the three dimensions of preparation, implementation process, and comprehensive evaluation of the teaching clinics. The response rates for the two rounds of the expert consultation were both 100.00%, with expert authority coefficients of 0.89 and 0.90, respectively. The overall Cronbach's α coefficient of the questionnaire was 0.93, and the correlation coefficients between each item score and the total score were all >0.30. Structural validity analysis revealed that three common factors were extracted from the questionnaire, with a cumulative variance contribution rate of 77.89%. Conclusions:The General Practice Teaching Clinic Quality Evaluation Questionnaire for Residents developed in this study demonstrates high reliability and validity. The questionnaire provides a scientific basis for the standardized assessment of teaching quality in general practice clinics. By incorporating resident feedback on the teaching process, the questionnaire promotes the development of a teaching clinic quality improvement mechanism focused on residents and plays a significant role in enhancing the teaching capabilities of supervising physicians in clinics.

Microfibrino-associated protein (MFAP) 2 is an extracellular matrix glycoprotein and a member of the MFAP family. It participates in the assembly of extracellular elastic microfibers.Upregulation of MFAP2 can promote the occurrence and development of various tumors and regulate multiple cancer-related signaling pathways and is related to their prognosis, making it a potential target for tumor treatment. This article summarizes the research progress on the pathogenesis, targeted therapy and prognosis of MFAP2 in malignant tumors.

Objective To address the challenges of accurately capturing critical features in small-sample diabetic retinopathy(DR)recognition models in real-world applications,and the overly smooth distribution of true and false feature coefficients,we propose an enhanced small-sample DR recognition method based on an improved capsule network.Methods Firstly,the method enhances image feature representation by removing redundant boundary information and employing discrete wavelet transform based on the Haar wavelet function,thereby high-lighting critical pathological features.Secondly,the convolutional layer of the capsule network is optimized through a multi-branch architecture to extract multi-scale features from retinal images,while incorporating a convolutional block attention module that is subsequently fed into the capsule layer.Finally,the sigmoid function replaces the soft-max function in dynamic routing,thereby improving the model's robustness.Result The enhanced neural network model achieved an accuracy of 98.62%on the Kaggle public dataset following a rigorous selection and preprocessing procedure.Conclusion The enhanced capsule network demonstrated superior precision in identifying diabetic reti-nopathy within small sample sizes compared to other state-of-the-art algorithms currently available.