Objective To construct prostate cancer(PCa)prediction models based on machine learning algorithms, so as to improve the accuracy of early diagnosis of PCa. Methods A retrospective analysis was performed on the clinical data of 504 patients who underwent prostate biopsy at our hospital during Jan. 2020 and Nov. 2024. Patients' age, body mass index(BMI), history of hypertension, diabetes and smoking, total prostate-specific antigen(tPSA), free prostate-specific antigen(fPSA), f/tPSA, prostate volume(PV), neutrophil count, lymphocyte count, neutrophil-to-lymphocyte ratio(NLR), Prostate Imaging Reporting and Data System(PI-RADS)score, digital rectal examination(DRE)results, and pathological findings were collected. The patients were divided into the training and testing sets at a ratio of 7:3. Ten early diagnosis prediction models of PCa were constructed using 10 supervised machine learning algorithms. Model performance was evaluated and validated using metrics including area under the receiver operating characteristic curve(AUC), accuracy, sensitivity, specificity, calibration curves, and decision curve analysis(DCA). SHAP analysis was used to interpret the models, and to clarify the importance of each feature and the basis for model decisions. Results All models showed good predictive value, with the gradient boosting machine(GBM)model performing the best(AUC=0.905, accuracy=84.1%, sensitivity=90.2%, specificity=80.0%). Calibration curves indicated good calibration and fitting of the GBM model, while DCA demonstrated favorable clinical net benefits. SHAP analysis identified the most significant features affecting PCa occurrence in descending order:tPSA, f/tPSA, PIRADS score, PV, age, and DRE. Conclusion The GBM model exhibits the optimal performance among the 10 models. The importance of features for predicting the occurrence of PCa, from the highest to the lowest, is as follows:tPSA, f/tPSA, PIRADS score, PV, age, and DRE.

Objective:To explore the predictive factors for ureteral stone impaction preoperatively and to construct a nomogram prediction model for impacted ureteral stones.Methods:A retrospective analysis was conducted on the clinical data of 209 patients with ureteral stones treated at The Second Affiliated Hospital of Zhengzhou University from July 2023 to June 2024. There were 164 males（78.5%）and 45 females（21.5%）. The age was 49（47，57）years，and the body mass index（BMI）was 25.10（23.55，27.24）kg/m2. Of the patients，85（40.7%）had comorbid hypertension and 85（40.7%）had comorbid diabetes. Stones were located on the left side in 124 patients（59.3%）and on the right side in 85 patients（40.7%）. Hydronephrosis was present in 169 patients（80.9%），and urine culture was positive in 29 patients（13.9%）. Patients were divided into impacted and non-impacted groups based on the presence or absence of ureteral stone impaction. Univariate and multivariate logistic regression analyses were performed to determine independent predictive factors for impacted ureteral stones. A nomogram model was constructed based on these results. The performance of the predictive model was evaluated using receiver operating characteristic（ROC）curves，calibration plots，and decision curve analysis（DCA）.Results:Among the 209 patients in this study，85（40.7%）experienced ureteral stone impaction. The impacted group had a significantly higher neutrophil-to-lymphocyte ratio（NLR）than the non-impacted group（3.91 ± 2.05 vs. 3.25 ± 2.10， P = 0.024），a higher rate of hydronephrosis［81.2%（69/85）vs. 80.6%（100/124）， P = 0.002］，larger stone surface area［（64.96 ± 39.96）mm2 vs.（51.86 ± 39.80）mm2， P = 0.021］，greater ureteral wall thickness（UWT）［（3.96 ± 1.37）mm vs.（3.06 ± 1.33）mm， P < 0.001］，and a higher ratio of the upper ureter diameter（D1）to the lower ureter diameter（D2）（DDR）（2.87 ± 1.58 vs. 2.00 ± 0.99， P < 0.001）. Univariate analysis showed that NLR，hydronephrosis，stone length，stone surface area，UWT，D1，D2，and DDR were statistically significant（ P < 0.05）. After multivariate logistic regression analysis，the following items were identified as independent predictors of impacted ureteral stones：NLR（ OR = 1.205，95% CI 1.026 - 1.415， P = 0.023），hydronephrosis（ OR = 1.840，95% CI 1.236 - 2.740， P = 0.003），stone length（ OR = 1.587，95% CI 1.142 - 2.206， P = 0.006），ureteral wall thickness（UWT）（ OR = 1.643，95% CI 1.263 - 2.136， P < 0.001），and DDR（ OR = 2.907，95% CI 1.040 - 8.130， P = 0.042）.Based on these independent predictive factors，a nomogram prediction model for impacted ureteral stones was constructed. The area under the ROC curve was 0.797（95% CI 0.737 - 0.858），and the calibration curve showed good consistency. The decision curve suggested that the model had good clinical net benefit. Conclusions:NLR，hydronephrosis，stone length，UWT，and DDR are all independent predictors for impacted ureteral stones. The nomogram model constructed based on these factors has good predictive performance.

BACKGROUND:Ligament rupture and defect of the lateral or medial malleolus caused by high-energy injuries are common challenges in foot and ankle surgery.Their neighboring core tendons are often used as grafts to reconstruct the deficient ligaments.It is of paramount importance to investigate the mechanical properties of such tendons in the context of ligament defects to provide a suitable donor tendon.OBJECTIVE:To investigate the interactive dynamics and biomechanical alterations among their core tendons during ankle joint motions under varying degrees of lateral or medial malleolar ligament defect.METHODS:Based on CT imaging data of the left foot of a 50-year-old healthy male,a surface stereolithography model was extracted and constructed using MIMICS software.After Geomagic Wrap software was employed to fit the surfaces,a bone-cartilage-ligament-tendon ankle complex model incorporating varying degrees of ligament deficiencies was created within SOLIDWORKS software.Finite element analysis was then conducted using Ansys Workbench software,and the model's validity was verified through a simulated anterior drawer test.Following validation,the mechanical response of the ankle under the conditions of internal and external rotation,as well as inversion and eversion,was simulated.The variation and distribution patterns of the maximum Von-Mises stress in the peroneus brevis and longus tendons,as well as the anterior and posterior tibial tendons,were observed.RESULTS AND CONCLUSION:(1)In the anterior drawer test,the maximum talar displacement reached 5.208 5 mm,which was similar to the data in the previous literature,thereby the effectiveness of the model was validated.(2)Under four loading conditions,the defect of unilateral single-bundle ligaments exerted minimal influence on the maximum stress of adjacent core tendons,whereas the defect of unilateral multi-bundle ligament significantly increased the maximum tendon stress.Except for the consistently high stress across segments of the anterior tibial tendon,the high-stress regions in the long and short peroneal tendons and posterior tibial tendon were concentrated at their distal ends near the insertions.(3)Regarding the defect of the lateral malleolar ligament,the maximum stress and its variation in the peroneus brevis tendon during inversion and internal rotation were higher than those in the peroneus longus tendon.During inversion under the condition of the defect of the anterior talofibular ligament,the maximum stress in the short peroneal tendon increased by 0.951 2 MPa compared to that of normal condition,while that in the long peroneal tendon decreased by 0.065 1 MPa.Under the condition of the defect of the calcaneofibular ligament during internal rotation,the maximum stress in the short peroneal tendon increased by 2.352 9 MPa,while the maximum stress in the long peroneal tendon decreased by 0.269 2 MPa.(4)During eversion and external rotation under the defect of medial malleolar ligament,the variations in the maximum stress of the anterior and posterior tibial tendons were complex and depended on the type of ligament defect.Notably,full-thickness ligament defect significantly augmented the maximum stress in both tendons.

Objective To analyze the characteristics of coronary atherosclerosis(CAS)and the gender differences in type 2 diabetes mellitus(T2DM)patients without cardiovascular disease symptoms using coronary artery computed tomography(CCTA).Methods A total of 178 asymptomatic individuals with T2DM who underwent CCTA examination due to CAS risk factors were included in this analysis in 2019,and they were divided into male group(n=105)and female group(n=73)according to sex.The general data and biochemical indicators of patients of different genders were compared.Logistic regression was used to analyze the influencing factors of high-risk signs of CAS in patients with T2DM.The incidences of coronary artery involvement score(SIS)≥4 and segmental stenosis score(St.S)≥3 in male and female patients with different courses of DM were compared.Results The age and the proportion of hypertension in female patients with T2DM were higher than those in male patients(P<0.05),while the body mass index(BMI),smoking history,triglyceride(TG),total cholesterol(TC)/high-density lipoprotein cholesterol(HDL-C),and proportions of SIS≥4 were lower than those in male patients(P<0.05).Logistic regression analysis showed that the hypertension duration was an influencing factor for the occurrence of high-risk signs of CAS in patients with T2DM.The proportion of SIS≥4 in the female DM duration was higher than<5 years(P<0.05),and the proportion of SIS≥4 in>15 years was lower than<5 years(P<0.05).Conclusions The incidence of CAS high-risk signs in T2DM patients without cardiovascular symptoms is different between different sexes,the number of affected segments in women was related to the DM duration.

This article is to evaluate the clinical outcomes and health economics of minimally invasive surgery for hypertensive intracerebral hemorrhage(HICH).This review systematically compares clinical efficacy and economic value of three minimally invasive techniques:small bone window microsurgery,neuroendoscopic surgery and stereotactic drainage based on 12 randomized controlled trials(RCTs),7 cohort studies,and 8 economic evaluations 2019-2024.Cost-effectiveness analysis(CEA)and cost-utility analysis(CUA)were employed to assess resource utilization and health outcomes.Minimally invasive approaches overall outperform conventional craniotomy.Stereotactic surgery achieves the shortest hospitalization(5-8 days)and lowest direct costs;neuroendoscopy significantly improves quality of life(quality-adjusted life years(QALYs);and small bone window surgery offers the best postoperative stability.It is recommended to choose the surgical method based on patient characteristics and to optimize healthcare resource allocation through medical insurance payment reform and technology promotion.

The knowledge graph has the characteristic of visualizing structural knowledge and has gradually become a research hotspot in the field of TCM diagnosis and treatment.This article reviewed the development trend and construction techniques of knowledge graph,summarized its application status in the field of TCM diagnosis and treatment,and provided a comprehensive review from the aspects of basic knowledge of TCM,inheritance of experience of renowned doctors,TCM question answering systems,and exploration of auxiliary decision-making.It also summarized the problems of knowledge graph technology in solving the informationization and intelligence process of TCM,and discussed and prospect the research and application directions of knowledge graph in the field of informationization and intelligence of TCM based on the characteristics of TCM knowledge.

Objective: To investigate the efficacy and safety of intelligent temperature-pressure-controlled flexible ureteroscopy combined with negative-pressure suction sheath lithotripsy in the treatment of upper urinary tract stones ≤2.5 cm. Methods: The clinical data of 225 patients with ≤2.5 cm upper urinary tract stones treated with this surgical method in our department during Aug. 2023 and Jul. 2024 were retrospectively analyzed. The patients were divided into the dual-control group (n=36) and conventional group (n=189) according to whether or not the intelligent temperature and pressure control device was used during operation. In the dual-control group，the intraoperative temperature and pressure in the renal pelvis were monitored and controlled in real time by the temperature and pressure sensors distributed at the end of the ureteral soft lens. The perioperative parameters，stone-removal rate，complication rate and renal function were compared between the two groups. Results: All operations were successfully completed in both groups. The postoperative procalcitonin (PCT) level [(22.75±5.85) ng/L vs. (29.08±6.60) ng/L，P=0.001]，difference in the white blood cell (WBC) level [(0.24±2.12)×10 cells/L vs. (1.19±2.17)×10 cells/L，P=0.016]，incidence of fever (2.8% vs. 16.9%，P=0.028) and overall complication rate (5.6% vs. 19.6%，P=0.042) were significantly lower in the dual-control group than in the conventional group，while the stone-clearance rate was slightly higher (88.9% vs. 82.5%，P=0.346)，with no significant difference. Conclusion: For upper urinary tract stones ≤2.5 cm，intelligent temperature-pressure-controlled ureteroscopy combined with negative-pressure suction sheath lithotripsy has a satisfactory stone-removal rate and a low rate of complications，which is worthy of clinical promotion.

ObjectiveThis study aims to develop and validate a novel multimodal predictive model， termed criss-cross network（CCNet）-directed graph neural network（DGNN）（CGN）， for accurate assessment of pulmonary nodule progression in high-risk individuals for lung cancer， by integrating longitudinal chest computed tomography（CT） imaging with both traditional Chinese and western clinical evaluation data. MethodsA cohort of 4 432 patients with pulmonary nodules was retrospectively analyzed. A twin CCNet was employed to extract spatiotemporal representations from paired sequential CT scans. Structured clinical assessment and imaging-derived features were encoded via a multilayer perceptron， and a similarity-based alignment strategy was adopted to harmonize multimodal imaging features across temporal dimensions. Subsequently， a DGNN was constructed to integrate heterogeneous features， where nodes represented modality-specific embeddings and edges denoted inter-modal information flow. Finally， model optimization was performed using a joint loss function combining cross-entropy and cosine similarity loss， facilitating robust classification of nodule progression status. ResultsThe proposed CGN model demonstrated superior predictive performance on the held-out test set， achieving an area under the receiver operating characteristic curve（AUC） of 0.830， accuracy of 0.843， sensitivity of 0.657， specificity of 0.712， Cohen's Kappa of 0.417， and F₁ score of 0.544. Compared with unimodal baselines， the CGN model yielded a 36%-48% relative improvement in AUC. Ablation studies revealed a 2%-22% increase in AUC when compared to simplified architectures lacking key components， substantiating the efficacy of the proposed multimodal fusion strategy and modular design. Incorporation of traditional Chinese medicine （TCM）-specific symptomatology led to an additional 5% improvement in AUC， underscoring the complementary value of integrating TCM and western clinical data. Through gradient-weighted activation mapping visualization analysis， it was found that the model's attention predominantly focused on nodule regions and effectively captured dynamic associations between clinical data and imaging-derived features. ConclusionThe CGN model， by synergistically combining cross-attention encoding with directed graph-based feature integration， enables effective alignment and fusion of heterogeneous multimodal data. The incorporation of both TCM and western clinical information facilitates complementary feature enrichment， thereby enhancing predictive accuracy for pulmonary nodule progression. This approach holds significant potential for supporting intelligent risk stratification and personalized surveillance strategies in lung cancer prevention.