ObjectiveTo address the challenges of unstructured classical Chinese expressions， nested entity relationships， and limited annotated data in famous traditional Chinese medicine（TCM） case records， this study proposes a joint relation extraction framework that integrates data augmentation and entity mapping， aiming to support the construction of TCM diagnostic knowledge graphs and clinical pattern mining. MethodsWe developed an annotation structure for entities and their relationships in TCM case texts and applied a data augmentation strategy by incorporating multiple ancient texts to expand the relation extraction dataset. A cascade binary tagging framework for relation triple extraction（CasRel） model for TCM semantics was designed， integrating a pre-trained bidirectional encoder representations from transformers（BERT） layer for classical TCM texts to enhance semantic representation， and using a head entity-relation-tail entity mapping mechanism to address entity nesting and relation overlapping issues. ResultsExperimental results showed that the CasRel model， combining data augmentation and entity mapping， outperformed the pipeline-based Bert-Radical-Lexicon（BRL）-bidirectional long short-term memory（BiLSTM）-Attention model. The overall precision， recall， and F₁-score across 12 relation types reached 65.73%， 64.03%， and 64.87%， which represent improvements of 14.26%， 7.98%， and 11.21% compared to the BRL-BiLSTM-Attention model， respectively. Notably， the F₁-score for tongue syndrome relations increased by 22.68%（69.32%）， and the prescription-syndrome relations performed the best with the F₁-score of 70.10%. ConclusionThe proposed framework significantly improves the semantic representation and complex dependencies in TCM texts， offering a reusable technical framework for structured mining of TCM case records. The constructed knowledge graph can support clinical syndrome differentiation， prescription optimization， and drug compatibility， providing a methodological reference for TCM artificial intelligence research.

OBJECTIVE To investigate the mechanism by which Shenfukang capsule （SFK） ameliorates renal interstitial fibrosis （RIF） by regulating macrophage-fibroblast crosstalk via the signal transducer and activator of transcription 3 （STAT3）/Yes-associated protein （YAP） pathway. METHODS RAW264.7 macrophages were induced to polarize with transforming growth factor-β 1 （TGF-β 1 ）. Subsequently， a non-contact co-culture system of macrophages with renal fibroblasts NRK-49F， as well as a non-contact co-culture system of macrophages with YAP-knockdown renal fibroblasts， were established. Cells were treated with low， medium， and high concentrations （4， 8， 16 μg/mL） of SFK， as well as a STAT3 inhibitor （STAT3-I，1.64 μg/mL） and losartan potassium tablets （positive control， 0.28 μg/mL）， for 48 h. After intervention， the protein and mRNA expression levels of CD86， CD163， and STAT3 in macrophages， as well as CD86， F4/80， and STAT3 mRNA， were detected. In co-cultured renal fibroblasts， the protein and mRNA expression levels of α -smooth muscle actin （ α -SMA）， Vimentin， collagen type Ⅰ （Col-Ⅰ）， matrix metalloproteinase-1 （MMP-1）， and YAP were detected. In co-cultured YAP-knockdown renal fibroblasts， the protein and mRNA expression levels of α -SMA， Vimentin， and MMP-1 were also detected. RESULTS Following TGF-β 1 induction， the protein expression levels of CD86， CD163， and STAT3， as well as the mRNA expression levels of CD86， F4/80， and STAT3 in macrophages were significantly increased （ P ＜0.05）. The conditioned medium from polarized macrophages activated renal fibroblasts， as evidenced by significantly increased protein and mRNA expression levels of α -SMA， Vimentin， Col-Ⅰ， and YAP， and significantly decreased protein and mRNA expression levels of MMP-1 in renal fibroblasts （ P ＜0.05）. Treatment with SFK and STAT3-I reversed the changes in the above indicators， with the medium concentration SFK group showing stronge r effects on some indicators than the low and high concentration groups. In the non-contact co-culture experiment of macrophages and YAP-knockdown renal fibroblasts， there were no statistically significant differences in the protein and mRNA expression of α-SMA， Vimentin， and MMP-1 among the groups. CONCLUSIONS SFK can inhibit macrophage-renal fibroblast crosstalk by blocking the STAT3/YAP pathway， thereby delaying the progression of RIF.

In classical medical literature， the concept of "toxin" in traditional Chinese medicine （TCM） encompasses medicinal properties， etiology and pathogenesis， syndromes， and treatment. Its connotations extend beyond the intrinsic potency and potential toxicity of medicinal substances to include pathological states characterized by the accumulation， latent persistence， transmission and transformation of pathogenic toxins， as well as the turbid stagnation. The concept also involves the nomenclature of toxin-related disorders and therapeutic strategies for the application and elimination of toxins. Previous studies have largely focused on individual medical classics， specific diseases， or drug toxicity， while different conceptual levels of toxin have often been conflated. After briefly tracing broader semantics of toxin， this paper adopts medical properties， etiology and pathogenesis， syndromes and treatment as the analytical framework to examine the historical evolution of its medical connotations across different historical periods， aiming to clarify the conceptual hierarchy of toxin in TCM and elucidate its developmental trajectory. The basic meanings of toxin took shape in the pre-Qin and Qin-Han dynasties， encompassing medicinal efficacy and risks， pathogenic qi leading to damage， disease and syndrome nomenclature， and the principles of using toxic substances to treat diseases. The understanding of accumulation， latent persistence， and transmission of toxin pathogen， as well as residual toxin was further enriched in the Jin-Sui-Tang dynasties. The materia medica standardization and syndrome-based treatment of toxin were witnessed in the Song-Jin-Yuan dynaties， while further elaboration of the toxin concept in warm epidemic， warm disease， and surgery were promoted during the Ming-Qing dyanties. In the modern and contemporary periods， more complex pathogenic concepts emerged， including collateral damage by toxin， toxin bi （痹）， and turbid toxin. These historical developments provide important documentary evidence for the standardization of related terminology， syndrome research， and the safety evaluation of medicinal applications.

Objective To quantitatively evaluate age-,sex-,and segment-related characteristics of lumbar disc degeneration(LDD)in patients with low back pain(LBP)using T1rho MRI,and assess its diagnostic value in early-stage disc degeneration.Methods A retrospective cross-sectional study was conducted on 89 LBP patients(balanced distribution across ages 20～29,30～39,40～49,50～59 years and genders)admitted in Department of Orthopedics of First Affiliated Hospital of Army Medical University from March 2024 to February 2025.All patients underwent lumbar MRI including T1rho sequences.Two senior radiologists performed blinded Pfirrmann grading on 445 intervertebral discs and measured T1rho values of the nucleus pulposus(NP).Kappa test was used to evaluate the inter-observer agreement of Pfirrmann grading,and Spearman correlation coefficient was employed to analyze the correlation between Pfirrmann grade and NP-T1rho value.Disc degeneration characteristics were compared across different ages,sexes,and spinal segments.After adjusting for body mass index(BMI)and gender,partial correlation analysis was used to assess the correlation between NP-T1rho value of each lumbar segment and age.Receiver operating characteristic(ROC)curve was plotted to analyze the diagnostic performance of T1rho quantification for early LDD in LBP patients.Results Good inter-observer agreement was observed in Pfirrmann grading across segments,with a Kappa coefficient of>0.70.The NP-T1rho values of all NP were in a strong negative correlation with Pfirrmann grades(r=-0.877,P<0.001).The NP-T1rho values were progressively decreased from Pfirrmann gradesⅠ to Ⅳ(P<0.001),with a steep decline between gradesⅡ and Ⅲ.The patients aged 20～29 and 30～39 years showed significantly higher NP-T1rho values in Pfirrmann grades Ⅰ and Ⅱ discs when compared with those aged 40～49 and 50～59 years(P<0.05).The males had significantly higher mean NP-T1rho values in grade I than the females(P=0.006).In the upper spinal segments(L1～L4),the males exhibited significantly higher mean NP-T1rho value than the females(P=0.025),and NP-T1rho value was decreased obviously with increase of age across all age groups(P<0.05).Except L5/S1 segments,NP-T1rho value and age in all spinal segments had significant negative correlations(r=-0.548～-0.349,P<0.001).The area under the curve(AUC)for distinguishing grade Ⅰ from Ⅱ discs was 0.824(95%CI:0.772～0.876),and for grades Ⅱ from Ⅲ was 0.978(95%CI:0.964～0.992).Conclusion T1rho MRI provides accurate quantitative assessment of early lumbar disc degeneration in LBP patients,with an NP-T1rho value of 79.87 ms serving as a discriminative threshold between grade Ⅱ and grade Ⅲ.

Objective To investigate the association between serum magnesium levels,serum phosphorus concentrations,vascular calcification,and cardiovascular disease mortality in maintenance hemodialysis patients.Methods This study enrolled 200 hemodialysis patients admitted to Nanjing Drum Tower Hospital from May 2020 to May 2022 as subjects,with an additional 200 healthy individuals from the same period selected as a control group.The biochemical indicators between the two groups were compared;their correlations were analyzed.Binary logistic regression was used to investigate the independent factors of serum magnesium and phosphorus levels in relation to vascular calcification and cardiovascular events in maintenance hemodialysis patients.ROC curve analysis was employed to assess the predictive value of serum magnesium and phosphorus for vascular calcification and cardiovascular events.Results The research group's patients exhibited significantly elevated levels of blood phosphorus,calcium-phosphorus product,iPTH,AACS,and 25-(OH)-VitD compared to the control group.In contrast,their blood magnesium and BMP-7 levels were notably lower than those of the control group,with statistical significance(P<0.05).Pearson correlation showed positive correlations between serum magnesium and serum calcium,phosphorus,calcium-phosphorus product,25-(OH)-VitD3,and BMP-7(r=0.385,0.183,0.141,0.131,0.458,P<0.05);between serum calcium and serum phosphorus,calcium-phosphorus product,iPTH,AACS,25-(OH)-VitD3,and BMP-7(r=0.318),correlation(r=0.318,0.311,0.098,0.170,0.277,0.485,P<0.05);between serum phosphorus and calcium-phosphorus product,iPTH,AACS,25-(OH)-VitD3(r=0.362,0.506,0.367,0.461,P<0.05);between calcium-phosphorus product and iPTH,AACS,25-(OH)-VitD3(r=0.542,0.373,0.434,P<0.05);between iPTH and AACS,25-(OH)-VitD3 showing positive correlations(r=0.553,0.616,P<0.05)and a negative correlation with BMP-7(r=-0.373,P<0.05);between AACS and 25-(OH)-VitD3 showing a positive correlation(r=0.402,P<0.05),and a negative correlation with BMP-7(r=-0.155,P<0.05),with statistically significant differences(P<0.05).Statistically significant differences were noted between the two groups in age,diabetes,serum magnesium,serum calcium,serum phosphorus,calcium-phosphorus product,25-(OH)-VitD3,and hs-CRP(P<0.05).Logistic regression analysis showed that age,serum magnesium,serum calcium,serum phosphorus,calcium phosphate product,25-(OH)-vitamin I were all risk factors for cardiovascular disease(CVD)mortality in maintenance hemodialysis(MHD)patients(P<0.05).ROC curve analysis showed that serum magnesium,serum calcium,and serum phosphorus had predictive areas under the curve(AUC)of 0.895,0.802,and 0.851 for CVD mortality in MHD patients,with sensitivities and specificities of 87.5%/98.7%,66.7%/90.8%,and 72.9%/100%,respectively.The combined prediction for CVD mortality in MHD patients showed an AUC of 0.921,with a sensitivity of 81.3%and specificity of 93.4%.Conclusion MHD patients exhibit low blood magnesium levels,elevated serum phosphorus concentrations,and increased calcium-phosphorus product,with complex correlations among these biomarkers.Age,magnesium,calcium,and phosphorus levels were all associated with CVD mortality.ROC curve analysis demonstrates that magnesium,calcium,and phosphorus,both individually and in combination,have high predictive value for CVD mortality risk.

OBJECTIVE To construct a predictive tool for surgical site infections(SSIs)in spinal surgery for Chi-nese population to provide evidence support for reducing SSIs.METHODS A systematic review of Chinese and English database literature was conducted for Meta-analysis to obtain pooled risk values for influencing factors,and a risk prediction scoring tool was constructed based on the logistic regression model.Patients who underwent spinal surgery and completed postoperative follow-up in a tertiary hospital in Beijing from Jan.to Dec.2021 were selected to validate the predictive effect of the tool.RESULTS The predictive model for SSIs in Chinese spinal sur-gery patients was Logit(P)=—3.47+0.63(age 60 years)+0.31 ×(patient with cardiovascular disease)+0.69 ×(rheumatoid arthritis)+1.07 ×(diabetes)+1.06 ×(operation duration＞3 h)+1.17 ×(preopera-tive albumin＜35 g/L)+0.71 ×(history of spinal surgery)+0.67 ×(carrying internal implants)+0.73 ×(blood transfusion).The total score of the predictive tool was 92,with a cutoff score of≥24.50 indicating high-risk individuals.The area under the curve was 0.733,with the sensitivity 58.30％and the specificity 79.60％.CONCLUSION The established predictive model for SSIs in Chinese spine surgery demonstrates good predictive performance and can be used as a reference assessment tool in clinical practice.

This study aims to simultaneously detect three antibiotic resistance genes(blaNDM,mcr-1 and cfr).A triplex fluorescence quantitative PCR method was established.Plasmids,primers and probes were designed and optimized.The method could specifically detect blaNDM,mcr-1 and cfr,but not other antibiotic resistance genes.The R2 of the standard curves of the three antibiotic re-sistance genes were all greater than 0.999,and the coefficients of variation were all lower than 1％.The lowest detection limits of the plasmids were 1 × 102 copies/μL.This method was used to de-tect 800 bacterial samples.The results showed that 32 samples contained mcr-1 gene,40 samples contained blaNDM gene,2 samples contained cfr gene,8 samples contained both mcr-1 and blaNDM genes.There were no samples carrying three antibiotic resistance genes detected.The results indica-ted that the triplex fluorescence quantitative PCR method established in this experiment had the advantages of high sensitivity,specificity and stability.It was suitable for rapid detection of blaNDM,mcr-1 and cfr antibiotic resistance genes in clinical practice.It provided a convenient and quick method basis for the detection of antibiotic resistance genes.

Objective:To develop a deep learning model based on 3D super-resolution reconstruction technology and to analyze its feasibility and effectiveness in predicting paraglottic space invasion in hypopharyngeal cancer.Methods:A retrospective study was conducted involving 382 patients with hypopharyngeal squamous cell carcinoma treated at Qilu Hospital of Shandong University between January 2014 and December 2020. The cohort included 364 males and 18 females, with a mean age of 62±7 years. Patients were divided into a training set ( n=300) and a test set ( n=82) based on enrollment time. A generative adversarial network was used to perform 3D super-resolution reconstruction on contrast-enhanced CT images, improving spatial resolution by 16 times. A 2.5D deep learning strategy was employed to construct Resnet-NR and Resnet-SR models based on conventional and super-resolution images, respectively, to predict whether the paraglottic space was invaded. Model performance was evaluated using receiver operating characteristic (ROC) curves and area under the curve (AUC). A multi-reader multi-case study was conducted to assess the impact of the artificial intelligence (AI) model on clinicians′ diagnostic capabilities. Results:The super-resolution model Resnet-SR achieved the highest accuracy in both the training set (AUC=0.87, 95% CI: 0.84-0.90) and the test set (AUC=0.88, 95% CI: 0.81-0.96), significantly outperforming traditional clinical indicators (T stage, N stage, tumor diameter, and pathological differentiation degree) (AUC range: 0.55-0.70, all P<0.05). In comparison, the conventional-resolution model Resnet-NR achieved AUCs of 0.81 (95% CI: 0.77-0.84, P=0.005) and 0.80 (95% CI: 0.71-0.89, P=0.184) in the training and test sets, respectively. Using Resnet-SR to assist clinical decision-making improved the diagnostic accuracy of junior physicians (AUC=0.793 without AI assistance vs. AUC=0.871 with AI assistance, P=0.012) and significantly reduced diagnosis time for clinicians of all experience levels (86.5 s without AI assistance vs. 82.5 s with AI assistance, t=2.01, P=0.032). Conclusion:This study successfully develops a deep learning model based on 3D super-resolution reconstruction technology, which can assist in preoperative prediction of paraglottic space invasion in hypopharyngeal cancer. The AI-assisted tool improves diagnostic accuracy for junior physicians and enhances diagnostic efficiency for clinicians across all experience levels.

Background and Aims:In recent years,with the continuous progress of systemic therapy,hepatic arterial infusion chemotherapy(HAIC)combined with immune checkpoint inhibitors and anti-angiogenic agents has demonstrated significant efficacy in the treatment of advanced hepatocellular carcinoma(HCC).However,direct comparisons between different immunotherapeutic targets,such as PD-1 and PD-L1 inhibitors,in terms of clinical benefit and safety remain limited.This study aimed to compare the efficacy and safety of HAIC plus bevacizumab and sintilimab(HAIC-BP1)versus HAIC plus bevacizumab and atezolizumab(HAIC-BPL)in advanced HCC.Methods:A retrospective analysis was conducted on 88 patients with advanced HCC who received first-line HAIC-BP1or HAIC-BPL at Sun Yat-sen University Cancer Center between January 2020 and December 2022.Progression-free survival(PFS),overall survival(OS),objective response rate(ORR),disease control rate(DCR),and adverse events(AEs)were compared between the two groups.Cox regression analysis was performed to identify prognostic factors affecting PFS.Results:A total of 47 patients were included in the HAIC-BP1 group and 41 patients in the HAIC-BPL group,with no statistically significant differences in baseline characteristics between the two groups(all P>0.05).The ORR(59.6%vs.65.9%)and DCR(72.3%vs.80.5%)did not significantly differ between the HAIC-BP1 group and the HAIC-BPL group(both P>0.05).After a median follow-up of 16.3 months,there were no significant differences in median OS(21.3 months vs.22.4 months)or median PFS(6.7 months vs.6.2 months)between the HAIC-BP1 group and the HAIC-BPL group(both P>0.05).The incidence of AEs was similar,and no treatment-related deaths occurred.Multivariate Cox regression analysis identified tumor diameter>10 cm as an independent adverse prognostic factor for PFS(HR=0.48,95%CI=0.27-0.83,P=0.009).Conclusion:Both HAIC-BP1 and HAIC-BPL demonstrated comparable efficacy and favorable safety profiles as first-line treatment options for advanced HCC.Tumor diameter>10 cm was an independent unfavorable prognostic factor for PFS,underscoring the importance of patient stratification in clinical decision-making.

Purpose To quantitatively compare the diffusion parameters of mono-and biexponential diffusion-weighted imaging models,and to obtain optimal sets of b-values in diffusion-weighted MRI for obtaining monoexponential apparent diffusion coefficient(ADC)close to perfusion-insensitive intravoxel incoherent motion(IVIM)model ADC(ADCIVIM)in identifying of pulmonary solid benign and malignant lesions.Materials and Methods IVIM was performed in 40 patients with solid nodule and masse in Xi'an Gaoxin Hospital from July 2021 to August 2022 using a 3.0T MR imager.Two experienced diagnostic radiologists subjectively evaluated the IVIM images.A single index model was used to calculate ADC values(ADC0-1 000,ADC20-1 000,ADC50-1 000,ADC80-1 000,ADC150-1 000,ADC300-1 000,ADC500-1 000,ADC300,500,1 000,ADC300,800,1 000,ADC300,500,ADC300,800 and ADC300,1 000).The reference standard ADCIVIM value were calculated using a double-exponential model.The physician's measurements between two physicians were measured.The malignant and benign groups were compared and receiver operator characteristic curve for all parameters were analyzed.Results The measurement consistency of ADC values under b value sets and ADCIVIM was very good,and the intraclass correlation coefficient was more significant than 0.75.The differences between ADCIVIM and ADC values in each b group were statistically significant(t=-6.016--2.500,all P<0.05).The area under the curve(AUC)of ADCIVIM was the largest(0.906),with an optimal threshold of 1.271×10-3 mm2/s,a sensitivity of 80.0%and a specificity of 93.0%.The diagnostic efficacy close to ADCIVIM were ADC300,800(AUC=0.891),ADC50-1 000(AUC=0.827)and ADC300,800,1 000(AUC=0.795),respectively.The optimal threshold of ADC300,800 was 1.140×10-3 mm2/s,the sensitivity and specificity were 80.0%and 93.7%,respectively.Conclusion Combining b-values 300 s/mm2 and 800 s/mm2 is recommended as routine scanning parameters for identifying the insensitive monoexponential ADC between benign and malignant solid pulmonary lesions.