OBJECTIVE Identify the incentive and constraint factors of artificial intelligence （AI） application in the pharmaceutical field， and promote the application of AI in the field of pharmacy. METHODS Based on the technology acceptance model （TAM）， technology-organization-environment （TOE） framework， and diffusion of innovation theory （DOI）， a TAM-TOE-DOI integrated framework was constructed through a four-stage research process of “theoretical review → dimension mapping → mechanism integration → proposition development”. Combining the analytical pathways of the above three theories in AI application in pharmacy with the integration mechanisms and core propositions of the TAM-TOE-DOI， literature review and deductive reasoning were employed to systematically identify the incentive and constraint factors of AI application in pharmacy from three levels：micro （TAM）， meso （TOE）， and macro （DOI）， and to propose optimization strategies. RESULTS & CONCLUSIONS At the micro level， the efficiency transformation and quality improvement brought by AI technology were the main incentive factors for perceived usefulness， while technological complexity and algorithmic opacity were the main constraint factors for perceived ease of use. At the meso level， the completeness of technological infrastructure， the strength of top management support and innovation climate， as well as external institutional pressure and competitive driving forces were the core incentive factors， whereas scarcity of organizational resources and talent shortage were the main constraint factors. At the macro level， relative advantage and observability were typical incentive factors， while technological complexity was a typical constraint factor. China’s health administration， medical insurance authorities， and other relevant departments should coordinate efforts at the macro， meso， and micro levels to advance AI application in pharmacy： optimizing human-computer interaction and implementing tiered training programs at the micro level； reinforcing organizational support systems and capacity building at the meso level； dismantling data barriers and building social trust at the macro level. Differentiated implementation pathways should be developed for medical institutions at different tiers.

Objective To analyze a case of violations by an occupational medical examination (OME) institution and to explore the key control points for the supervision and management of OME institutions, as well as the core role of quality assessment in this context. Methods An OME institution suspected of illegal activities was used as the study subject. Retrospective analysis was conducted. Clues of suspected violations were identified by an on-site quality assessment. After investigation and verification by the local health authorities, legal action was taken against the institution for its violations. Results During an on-site quality assessment, the Guangdong Province OME quality control expert group discovered that the OME institution violated regulations, including unqualified personnel file, exceeding the scope of services category, issuing false reports, failing to report suspected occupational diseases on time, and failing to notify workers about suspected occupational diseases as required. The evidence was then submitted to the Guangdong Province OME Quality Control Center, which subsequently forwarded the case to local health administration department for filing and investigation. After the investigation, penalties were imposed on the OME institution for its illegal activities. Conclusion The key supervision and inspection points in the quality assessment of OME institutions include personnel file configuration, the quality control management system and its implementation, the quality of OME reports, and information reporting. Quality assessment plays a pivotal role in ensuring the legal and compliant practice of OME institutions, safeguarding the health rights and interests of workers, and enhancing the overall standard of the OME industry.

Objective To analyze ancient prescriptions for knee osteoarthritis(KOA),and explore their application patterns,for providing clinical reference value.Methods Prescriptions were collected from ancient documents,and a database was created by using Excel 2016 for frequency analysis.Association rules and hidden structures were analyzed by using SPSS Modeler 18.0 and Lantern 5.0.Results A total of 373 prescriptions and 421 herbs were collected,with 3894 times of frequency of medication.Danggui was the most frequently used herb.Herbs primarily tonified deficiencies,relieved exterior symptoms,and expelled wind-dampness,with warm properties and pungent,sweet,or bitter tastes,mainly targeting the spleen and kidney meridians.Thirteen strong association rules were identified,including Chuanxiong+Niuxi-Danggui,and Bixie-Niuxi.Hidden structure analysis revealed 10 variables,leading to 4 clusters and 4 core prescriptions.Common syndromes included wind-damp,wind-cold-damp,wind-damp-heat,and qi-blood deficiency.Conclusion The treatment of KOA in ancient literature focuses on dispelling wind and removing dampness,taking into account both positive deficiency and evil excess,flexible use of tonifying spleen and kidney,clearing dampness-heat,tonifying Qi and blood and other drugs.

Objective:To explore the differences in sociodemographic and clinical characteristics between pa-tients with unipolar depression bipolar depression and to establish a nomogram for identifying bipolar depression.Methods:Using data from the China Mental Disorders Cohort Study,the sociodemographic and clinical characteristics of 2 643 patients with unipolar depression and 250 patients with bipolar depression diagnosed accord-ing to the criteria of the Diagnostic and Statistical Manual of Mental Disorders,Fifth Edition(DSM-5)were includ-ed to compare their sociodemographic and clinical characteristics.These characteristics included general demograph-ic information,disease-related information,clinical examination results,and the severity of the disease assessed with the Global Assessment of Functioning(GAF)and Hamilton Depression Rating Scale.Logistic regression analysis was employed to identify factors influencing bipolar depression,and a nomogram was constructed for its identifica-tion.Results:The risk factors for bipolar depression included being male(OR=1.48),being employed(OR=1.38),having non-melancholic features during episodes(OR=2.33),a Body Mass Index ranging from normal to obese(OR=2.48,2.49,4.65),psychotic features(OR=2.14),mixed episode(OR=9.36),comorbid physical diseases(OR=2.47),four or more depressive episodes(OR=1.67),earlier age of onset(OR=0.95),longer ill-ness duration(OR=1.03),and higher GAF scores(OR=1.02).The nomogram model achieved an AUC of 0.81(95％CI:0.78-0.84).The Hosmer-Lemeshow test result was x2=6.96(P＞0.05),indicating good model fit.The calibration curve showed good performance.The decision curve analysis revealed that the nomogram pro-vides significant clinical benefit when the risk of bipolar depression was within the range of 0 to 0.9.Conclusion:The nomogram established based on the identified sociodemographic and clinical factors can accurately assess the risk of bipolar depression,providing a useful tool for early identification and intervention.

Objective:To investigate the correlation between serum growth differentiation factor 15 (GDF15) and the clinicopathological characteristics of patients with IgA nephropathy (IgAN), and further explore the relationship of GDF15 with the cardiac and renal prognosis of IgAN patients.Methods:It was a single-center retrospective cohort study. From January 2018 to December 2022, the relevant data were collected from patients who were diagnosed with primary IgAN at the Department of Nephrology, Beijing Anzhen Hospital Affiliated to Capital Medical University, and regularly followed up for at least 1 year. Serum samples were collected at admission and the baseline level of serum GDF15 was measured. Based on the median GDF15 level, IgAN patients were categorized into high-level GDF15 group and low-level GDF15 group, and their clinicopathological characteristics were compared. A multiple linear regression model was then constructed to identify independent factors associated with serum GDF15 level based on these comparisons. Subsequently, Kaplan-Meier survival analysis was performed to investigate the association between serum GDF15 level and the cardiorenal prognosis of IgAN patients.Results:A total of 104 IgAN patients were included in this study. The serum GDF15 level in these IgAN patients was 825.60 (556.84, 1 428.15) ng/L. Serum GDF15 level was positively correlated with 24 h urinary protein ( r=0.405, P<0.001), negatively correlated with estimated glomerular filtration rate (eGFR)( r=-0.606, P<0.001). The serum levels of GDF15 in patients with tubular atrophy or interstitial fibrosis (overall comparison among T0, T1, and T2, H=21.866, P<0.001), crescentic lesions (overall comparison among C0, C1, and C2, H=13.787, P=0.001), or intrarenal arteriolar lesions (overall comparison among none, mild, and moderate-to-severe, H=9.856, P=0.007) were significantly different. Compared with IgAN patients without tubular atrophy or interstitial fibrosis, those with Oxford classification T1 ( Z=-17.326, P=0.042) or T2 ( Z=-42.933, P<0.001) had higher serum GDF15 levels. Compared with IgAN patients without crescentic lesions, those with Oxford classification C2 had higher serum GDF15 levels ( Z=-45.929, P=0.001). Compared with IgAN patients without intrarenal arteriolar lesions, those with moderate-to-severe arteriolar sclerosis had higher serum GDF15 levels ( Z=-26.686, P=0.005). The median GDF15 was used as the cut-off value to divide IgAN patients into a high-level GDF15 group (≥825.60 ng/L, n=52) and a low-level GDF15 group (<825.60 ng/L, n=52). Compared to low-level GDF15 group, IgAN patients in high-level GDF15 group presented with a higher proportion of diabetes mellitus ( χ 2=9.420, P=0.002) and cardiovascular disease ( χ 2=7.792, P=0.005), a higher level of systolic blood pressure ( Z=-2.266, P=0.023), body mass index ( Z=-2.183, P=0.031), 24 h urinary protein ( Z=-3.485, P<0.001), blood total cholesterol ( Z=-2.002, P=0.045) and left ventricular mass index ( Z=-2.649, P=0.008), and a lower level of blood albumin ( Z=-3.053, P=0.002) and eGFR ( Z=6.480, P<0.001). Multiple linear regression analysis showed that serum GDF15 level was independently associated with systolic blood pressure (regression coefficient B=29.453, 95% CI 14.139–44.767, P<0.001), blood albumin ( B=-81.412, 95% CI -113.084–-49.740, P<0.001) and eGFR ( B=-9.797, 95% CI -17.554–-2.040, P=0.014). Moreover, IgAN patients in high-level GDF15 group exhibited significantly poorer cardiac and renal prognosis compared to low-level GDF15 group ( χ 2=9.955, P=0.002). Conclusion:High serum GDF15 level correlates with disease severity in IgAN, and high serum GDF15 level may suggest a poorer cardiorenal prognosis in IgAN patients.

Oral cancer is a common inflammation-associated malignant tumor of the head and neck.In recent years,the role of the tumor microenvironment in the progression of oral cancer has received increasing attention.In the microenvironment of solid tumors,including or-al cancer,tumor-associated macrophages(TAMs)are among the most common type of immune cell and are closely associated with the poor prognosis of more than 80%of human malignant tumors.An increasing body of evidence indicates that TAMs play important roles in the oc-currence and progression of oral cancer.On the basis of the most recent research progress,both domestic and international,we systematic-ally reviewed the phenotypic characteristics,clinicopathological correlations,and molecular regulatory mechanisms of TAMs in the microen-vironment of oral squamous cell carcinoma(OSCC).In particular,we focus on the key signaling pathways promoting tumorigenesis and de-velopment and immune escape mechanisms,with the aim of constructing a systematic theoretical framework for developing therapeutic strategies targeting TAMs and providing a scientific basis for clinical translational research.

Objective:To evaluate effectiveness of finerenone in patients with type 2 diabetes mellitus (T2DM) and chronic kidney disease (CKD) in the real world, and to analyze the associated factors of renal function progression during treatment.Methods:It was a single-center retrospective study. The patients diagnosed with T2DM and CKD who received finerenone treatment for 3 months in Beijing Anzhen Hospital, Capital Medical University between April 1 and October 1, 2023 were included. The clinical data before and after finerenone treatment were collected. Based on urinary albumin-to-creatinine ratio (UACR) and estimated glomerular filtration rate (eGFR), the patients were divided into different groups, and the differences of clinical data before and after treatment were compared respectively. Logistic regression models was used to analyze the correlated factors of renal function changes during the treatment.Results:A total of 151 patients were included with age of 63 (54, 70) years, and 103 males accounted for 68.2%. UACR level after 3 months of finerenone treatment was significantly lower than those before treatment ( Z=-5.051, P<0.001), whereas there was no statistically significant change in eGFR ( P>0.05). Both patients with baseline eGFR ≥60 ml·min -1·(1.73 m 2) -1 ( Z=-4.543, P<0.001) and those with baseline eGFR <60 ml·min -1·(1.73 m 2) -1 ( Z=-2.610, P=0.009) showed significant reductions in UACR after treatment. Both patients with baseline UACR ≥300 mg/g ( Z=-4.681, P<0.001) and those with baseline UACR <300 mg/g ( Z=-1.979, P=0.048) exhibited significantly lower UACR levels after treatment. The percentage reduction in UACR was greater in patients with baseline UACR ≥300 mg/g than in those with baseline UACR <300 mg/g ( Z=-2.102, P=0.036).After 3 months of finerenone therapy, serum potassium level was slightly higher than baseline, but the difference was not statistically significant ( P>0.05).The incidence of hyperkalemia after treatment was higher than baseline in patients with baseline eGFR <60 ml·min -1·(1.73 m 2) -1 ( χ 2=2.558 , P=0.039). During the treatment, 74 patients (49.0%) experienced renal function progression. Multivariate logistic regression analysis identified increased baseline serum albumin <45 g/L was an independent correlated factor of renal function progression during finerenone therapy ( OR=1.934, 95% CI 1.157-3.231, P=0.012). Conclusions:UACR in patients with T2DM and CKD can be reduced significantly after short-term treatment of finerenone. Increased baseline serum albumin level <45 g/L is independently associated with renal function progression during finerenone therapy.

Objective:To develop and validate a deep learning-based multimodal model integrating clinical and radiomic features for predicting acute kidney injury（AKI）risk after partial nephrectomy.Methods:A retrospective analysis was conducted on 416 patients who underwent partial nephrectomy at Zhongshan Hospital，Fudan University from January 2023 to January 2025. The cohort included 100 AKI patients［defined by a ≥ 25% reduction in postoperative evaluated glomerular filtration rate（eGFR）within 48 hours sustained for >24 hours］and 316 non-AKI patients（1∶3 ratio，randomly matched with 16 additional cases for redundancy）. Clinical and radiomic features were extracted from preoperative contrast-enhanced CT scans using PyRadiomics. Demographics included 259 males and 158 females，with a median age of 57（49，65）years，body mass index of（24.1 ± 3.3）kg/m2，preoperative eGFR of（88.5 ± 18.3）ml/（min·1.73 m2），postoperative eGFR（48-hour）of（76.0 ± 21.9）ml/（min·1.73 m2），Zhongshan Score（ZSscore）of 7.34 ± 2.01，and R.E.N.A.L. score of 7.50 ± 1.71. All tumors were T 1a stage. Patients were divided into training（n = 312）and test（n = 104）sets（3∶1 ratio）. A clinical model was constructed via multivariate logistic regression，while radiomic and combined（clinical + radiomic）models utilized an artificial neural network（ANN）with 1 input layer，5 hidden layers，1 output layer，and 10 5 training epochs. Model performance was evaluated by using receiver operating characteristic（ROC）curves and area under the curve（AUC），and was compared to the Martini model. Feature contributions were interpreted via SHapley Additive exPlanations（SHAP）. Results:In the test set，the results of multivariate logistic regression showed that patient’s weight，preoperative eGFR，R.E.N.A.L. score，surgical approach，and operation time were risk factors for AKI（ P < 0.05）. The AUC of the clinical feature prediction model constructed based on the above factors was 0.852（95% CI 0.775?0.929）. In the test set，the AUC of the Martini model was 0.725（95% CI 0.565?0.791）. The radiomic model，trained on 1 315 imaging features，achieved an AUC of 0.898（95% CI 0.804?0.993）with 94.2%（98/104）accuracy. The combined clinical and radiomic model，integrating 1 315 radiomic features and clinical features，demonstrated superior performance with an AUC of 0.946（95% CI 0.887?1.000）and 96.2%（100/104）accuracy，outperforming both the clinical model（ P = 0.03）and the Martini model（ P < 0.01）. SHAP analysis identified the top five predictors in the combined model：ZSscore（SHAP value：0.78），long-run low gray-level emphasis（SHAP value：0.61），run-length non-uniformity（SHAP value：0.58），size-zone non-uniformity（SHAP value：0.46），and gray-level co-occurrence matrix joint energy（SHAP value：0.36）. Conclusions:The deep learning-based multimodal model integrating clinical and radiomic features accurately predicts AKI risk after partial nephrectomy，offering a novel strategy for preoperative risk stratification and personalized intervention.

Objective:To explore the value of ultrasound imaging characteristics combined with clinical indicators in assessing the prognosis of patients with pancreatic ductal adenocarcinoma (PDAC).Methods:A retrospective analysis was conducted for patients who underwent pancreatic contrast-enhanced ultrasound (CEUS) from September 2017 to October 2023 at Peking Union Medical College Hospital and were diagnosed with PDAC based on pathological findings. Various parameters were recorded, including CA19-9 levels, tumor size, location, morphologic features, echogenicity, presence of internal cystic components, dilatation of the main pancreatic duct, peripheral vascular invasion, CEUS characteristics, presence or absence of liver metastasis, and treatment methods. In April 2024, patient survival information was obtained through telephone follow-up or review of medical records. Based on the results of the cox regression model analysis, a nomogram model of the risk of death was developed. The receiver operating characteristic (ROC) curves were applied to evaluate the predictive efficacy of the model. The calibration curves were plotted to evaluate the accuracy of the model, and clinical decision curves were used to evaluate the clinical benefit of the model.Results:This study included a total of 207 patients with PDAC. As of April 2024, 71 patients were alive and 136 died, with a median survival time of 14 months (95% CI: 12 -17). Multivariate analysis confirmed that the elevated CA19-9 ( HR=1.689, 95% CI: 1.102-2.588), tumor size >4 cm ( HR=1.641, 95% CI: 1.159-2.322), taller-than-wide shapes ( HR=1.450, 95% CI: 1.019-2.065), incomplete hypo-enhancement ( HR=1.618, 95% CI: 1.100-2.380), and liver metastasis ( HR=1.687, 95% CI: 1.175-2.423) were independent risk factors for survival in patients with PDAC. A nomogram model was further constructed for 6-month, 12-month and 3-year survival of patients with PDAC. The areas under the ROC curve were 0.679, 0.705 and 0.815, respectively. The calibration curves suggested that the model was more accurate, and the clinical decision curves showed that the model had a better clinical benefit. Conclusion:The combined use of ultrasound imaging characteristics and clinical indicators could effectively predict the prognosis of PDAC patients. Specifically, tumor size >4 cm, taller-than-wide shapes, incomplete hypo-enhancement, elevated CA19-9, and the presence of liver metastasis are correlated with poorer survival outcomes. The nomogram model constructed on the basis of these factors can be used to assess the survival of patients with PDAC.

The pharmaceutical industry faces challenges in quality digitization for complex multi-stage processes, especially in small-sample systems. Here, an intelligent quality prediction and diagnostic (IQPD) framework was developed and applied to Tong Ren Tang's Niuhuang Qingxin Pills, utilizing four years of data collected from four production units, covering the entire process from raw materials to finished products. In this framework, a novel path-enhanced double ensemble quality prediction model (PeDGAT) is proposed, which combines a graph attention network and path information to encode inter-unit long-range and sequential dependencies. Additionally, the double ensemble strategy enhances model stability in small samples. Compared to global traditional models, PeDGAT achieves state-of-the-art results, with an average improvement of 13.18% and 87.67% in prediction accuracy and stability on three indicators. Additionally, a more in-depth diagnostic model leveraging grey correlation analysis and expert knowledge reduces reliance on large samples, offering a panoramic view of attribute relationships across units and improving process transparency. Finally, the IQPD framework integrates into a Human-Cyber-Physical system, enabling faster decision-making and real-time quality adjustments for Tong Ren Tang's Niuhuang Qingxin Pills, a product with annual sales exceeding 100 million CNY. This facilitates the transition from experience-driven to data-driven manufacturing.