Objective To develop and validate a machine learning model based on preoperative clinical characteristics, laboratory indices, and radiological features for the non-invasive prediction of spread through air spaces (STAS) in patients with early-stage lung adenocarcinoma. Methods Preoperative data from patients with early-stage lung adenocarcinoma who underwent surgical resection at Northern Jiangsu People's Hospital between January 2020 and August 2025 were retrospectively collected. The data included clinical characteristics, laboratory indices, and radiological features. Patients were divided into a STAS-positive and a STAS-negative group based on postoperative pathological findings. The dataset was randomly split into a training set and a testing set at a 7 : 3 ratio. Feature variables were selected using the maximum relevance and minimum redundancy (mRMR) algorithm and the least absolute shrinkage and selection operator (LASSO) regression. Five machine learning models were constructed: logistic regression (LR), random forest (RF), support vector machine (SVM), light gradient boosting machine (LightGBM), and extreme gradient boosting (XGBoost). Model performance was evaluated using the area under the receiver operating characteristic curve (AUC) and decision curve analysis (DCA). The shapley additive explanations (SHAP) method was employed to interpret the optimal prediction model. Results A total of 377 patients were included, comprising 177 (46.9%) males and 200 (53.1%) females, with a mean age of (63.31±9.73) years. There were 261 patients in the training set and 116 patients in the testing set. In the training set, statistically significant differences were observed between the STAS-positive group (n=130) and STAS-negative group (n=131) across multiple features, including age, sex, neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), clinical T stage, and maximum solid component diameter (P<0.05). A final set of 10 feature variables was selected by combining mRMR and LASSO regression, and five machine learning models (LR, RF, SVM, LightGBM, XGBoost) were developed. The XGBoost model demonstrated optimal performance in both the training and testing sets, achieving AUCs of 0.947 [95%CI (0.920, 0.975)] and 0.943 [95%CI (0.894, 0.993)], respectively. DCA indicated that the XGBoost model provided a high net clinical benefit across a wide range of threshold probabilities. SHAP analysis revealed that the vessel convergence sign, clinical T stage, age, consolidation-to-tumor ratio (CTR), and MLR were the features with the highest contributions to STAS prediction. Conclusion The XGBoost model effectively predicts preoperative STAS status in early-stage lung adenocarcinoma, exhibiting excellent discriminative performance and good clinical interpretability. Key predictors such as the vessel convergence sign, clinical T stage, age and CTR provide a crucial reference for preoperative risk assessment and the individualized selection of surgical strategies, ultimately benefiting patients.

Objective: To analyze the consumption frequency of major foods among Chinese children aged 6-17 years old, and to provide a basis for optimizing the dietary structure of children in China. Methods: Using data from the China Nutrition and Health System Survey and Application Program for Children 0-18 years old, 56 734 children aged 6-17 years old from North, Norththeast East, Central, South, Southwest and Northwest seven regions in China were selected for the study using stratified cluster random sampling from 2019 to 2021. A food frequency questionnaire was used to investigate the intake frequency of eight food groups in a month, including fresh vegetables, fresh fruits, livestock and poultry meats, aquatic products, eggs, dairy products, legumes, and cereals and potatoes. The foods were grouped according to whether they met the recommended intake criteria outlined in the Dietary Guidelines for Chinese Residents 2022. The〖KG*2〗χ2 test was used to compare the differences in the proportion of childrens intake frequency of each food group meeting the standard in different regions and age groups. Results: The proportions of Chinese children aged 6-17 years who consumed fresh vegetables and cereals and potatoes ≥3 times/d were 12.1% and 67.2%, respectively. The proportions of children who consumed fresh fruits, livestock and poultry meats, eggs and dairy products ≥1 time/d were 50.8%, 58.8%, 36.0% and 54.3%, respectively. The proportion of legumes consumed ≥4 times/week was 37.4%, and the proportion of aquatic products consumed ≥2 times/week was 39.7%. Fresh vegetables (5.5%), fresh fruits (33.1%), and dairy products (36.4%) had the lowest frequency of meeting the recommended standards in South China, and aquatic products (27.4%) and eggs (21.1%) had the lowest frequency of meeting the recommended standards in Northwest (P<0.008 3). Conclusion The overall intake frequency of fresh vegetables, fresh fruits, legumes, and dairy products are insufficient among Chinese children, with significant regional variations.

Objective To analyze the relationship of total imaging burden score with blood pressure variability(BPV)and cognitive function in elderly patients with cerebral small vessel disease(CSVD).Methods Clinical data of 182 elderly CSVD patients admitted in our hospital from December 2022 to January 2024 were collected and analyzed retrospectively.According to the results of Chinese Guidelines for Diagnosis and Treatment of CSVD-Related Cognitive Dysfunc-tion(2019),Montreal Cognitive Assessment(MoCA)and other tools for their cognitive impair-ment,they were divided into cognitively impaired group(76 cases)and cognitively normal group(106 cases).The total burden score and BPV indicators[24 h systolic blood pressure coefficient of variation(24 h SBPCV),24 h diastolic blood pressure coefficient of variation(24 h DBPCV)]were compared between the two groups.ROC curve was plotted to evaluate the diagnostic value of total burden score,24 h SBPCV and 24 h DBPCV on cognitive impairment in elderly CSVD patients.Multivariate logistic regression analysis was used to identify the risk factors of cognitive impairment in the patients.Based on the MoCA score of the cognitively impaired group,these pa-tients were further assigned into mild and moderate-to-severe cognitive impairment subgroups[with a MoCA score of 18-25(43 cases)and＜18(33 cases),respectively].Then the total burden score,24 h SBPCV and 24 h DBPCV were compared between the two subgroups.Pearson correla-tion coefficient was adopted to evaluate the correlation of severity of cognitive impairment with total burden score,24 h SBPCV and 24 h DBPCV in the elderly CSVD patients.Results The total burden score,24 h SBPCV and 24 h DBPCV were significantly higher in the cognitively impaired group than the cognitively normal group(P＜0.01).ROC curve analysis indicated that total bur-den score(AUC=0.953,95％CI:0.926-0.980,P=0.000),24 h SBPCV(AUC=0.850,95％CI:0.795-0.906,P=0.000)and 24 h DBPCV(AUC=0.761,95％CI:0.690-0.832,P=0.000)had good diagnostic efficiency for cognitive impairment in the elderly CSVD patients,with a cut-off value of 1.5,11.82％,and 8.92％,respectively.Multivariate logistic regression analysis revealed that the above three indicators were risk factors for cognitive impairment in the elderly patients with CSVD(P＜0.05,P＜0.01).Their values were significantly lower in the mild than the moder-ate-to-severe cognitive impairment subgroups(P＜0.01).Pearson correlation analysis displayed that MoCA score was negatively correlated with total burden score,24 h SBPCV and 24 h DBPCV in elderly patients with CSVD(r=-0.755,-0.632,-0.601,P＜0.01).Conclusion Detection of total burden score and BPV indicators is beneficial to the assessment of cognitive impairment in elderly CSVD patients.The higher the total burden score and the greater the BPV indicators,the more severe the cognitive impairment is,which may explore new ideas for clinical diagnosis and treatment of CSVD.

Objective:To establish a risk prediction model by analyzing the influencing factors of delivery be-fore 34 weeks' gestatior with singleton pregnancies who underwent transcervical McDonald's cervical cerclage.Methods:Clinical data of 500 pregnant women with singleton pregnancies who underwent McDonald's cervical cerclage in the Department of Obstetrics and Gynecology,Heping Hospital,General Hospital of the People's Lib-eration Army Northern Theater of Operations from January 1,2019 to February 29,2024 were retrospectively ana-lyzed,and were randomly divided into a training set of 350 cases versus a validation set of 150 cases in accord-ance with a 7∶3 randomization.The data in the training set were screened for predictive variables affecting deliv-ery before 34 weeks' gestation after cervical cerclage using univariate combined multivariate Logistic regression,predictive modeling and assessment were established,and internal validation was performed in the validation set.Results:Preoperative cervical length(CL)＜20 mm,amniotic fluid deposits(AFS),absence of Lactobacillus vagi-nalis,and cervical dilatation＞3 cm were independent risk factors for adverse outcomes of labor and delivery be-fore 34 weeks' gestation after cervical cerclage(OR＞1,P＜0.05),and the Nomogram model constructed on the basis of this was found to have an AUC of 0.887,with a good agreement of the model(H-L test:P=0.905,Brier score of 0.095).The results of the clinical decision curve showed that the clinical decision value of the model was high when the threshold was between 0.10-0.92.The model was internally validated with an AUC of 0.812,and there was a high degree of agreement between the predictions and the actual situation(H-L test:P=0.859).Conclusions:The column-line diagram model constructed on this basis has a high clinical value,which can help us to screen out the high-risk pregnant women and give them individualized interventions in a timely manner.

Objective To analyze the application value of 3D Slicer software in endoscopic transsphenoidal surgery for pituitary adenoma resection.Methods A retrospective analysis was made on the clinical data of 36 patients with pituitary adenomas treated with 3D Slicer-assisted endoscopic transsphenoidal surgery(3D Slicer group)in the Department of Neurosurgery,The Second Affiliated Hospital of Xi'an Jiaotong University,from January 2024 to December 2024.Preoperatively,multimodal images were fused and reconstructed using 3D Slicer software to systematically evaluate bony anatomical structures such as sphenoid sinus ostia,intrasphenoidal septa,and sellar floor structures,design the size of pedicled nasoseptal flaps,and clarify the positional relationships between pituitary adenomas and surrounding vital structures including the internal carotid artery,pituitary gland,and optic chiasm,so as to provide real-time guidance for intraoperative procedures.Meanwhile,45 patients with pituitary adenomas treated with neuronavigation-assisted endoscopic transsphenoidal surgery from January 2023 to December 2023 were included as the control group(neuronavigation group).The surgical outcomes and postoperative complications were compared between the two groups.Results Compared with the neuronavigation group,the 3D Slicer group demonstrated higher identification rates of the optic nerve groove and carotid artery impression(94.4％vs.77.8％),shorter operative time[(2.9±0.6)h vs.(3.5±0.9)h],less intraoperative bleeding[(159.7±70.5)mL vs.(237.8±96.0)mL],and a lower incidence of postoperative olfactory dysfunction(8.3％vs.26.7％),with statistically significant differences(P＜0.05).However,there were no statistically significant differences between the two groups in the identification rate of the sphenoid sinus ostium(100.0％vs.97.8％),gross total resection rate(75.0％vs.64.4％),and the incidence of other postoperative complications,including cerebrospinal fluid leakage(0.0％vs.6.7％),intracranial infection(2.8％vs.11.1％),transient diabetes insipidus(30.6％vs.22.2％),and hypopituitarism(38.9％vs.37.8％,P＞0.05).Conclusion 3D Slicer software helps improve the mastery of anatomical basics in endoscopic transsphenoidal approach among junior and primary physicians,enhancing the clinical efficacy and safety of pituitary adenoma resection,and thus is worthy of clinical promotion.

Objective:To report the clinical, electrophysiological, and genetic mutation characteristics of a family with paramyotonia congenita, to enhance clinical understanding of ion channelopathies, provide better guidance for clinical diagnosis, and optimize treatment plans.Methods:The clinical data, nerve conduction studies, needle electromyography, long exercise test, short exercise test, and whole-exome sequencing results of the proband with paramyotonia congenita and his family members admitted to Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, in August 2023 were collected and analyzed.Results:The proband was a 15-year-old male who presented with episodic muscle weakness for 1 year, accompanied by elevated creatine kinase during episodes. Between the ages of 3 and 4, he exhibited facial and bilateral hand myotonia triggered by cold exposure. His mother, sister, maternal grandmother, maternal aunt and uncle also displayed similar cold-induced facial and hand myotonia during childhood. Additionally, his cousin exhibited walking difficulties, frequent falls, mild difficulty in opening eyes after closure, and mild chewing and swallowing difficulties at the age of 1. Electromyography of the proband showed widespread myotonic discharges and myopathic impairment, while the electromyography of his mother and sister revealed only widespread myotonic discharges. The proband′s long exercise test results were consistent with periodic paralysis, while the short exercise test results aligned with paramyotonia congenita. His sister′s long exercise test results were within normal range, while the short exercise test results of his mother and sister were consistent with paramyotonia congenita. Genetic testing of the family revealed that all tested patients carried a heterozygous SCN4A mutation, c.4343G>A (R1448H). Conclusions:The R1448H mutation in the SCN4A gene is the genetic cause of this family. Patients carrying the SCN4A R1448H mutation may exhibit clinical features of cold-induced myotonia with or without episodic muscle weakness. In addition to widespread myotonic discharges on electromyography, certain patients may also show myopathic impairment. The long and short exercise tests can further assist in differentiating various types of myotonia and periodic paralysis for clinical diagnosis.

Objective:To investigate the impact of receptor-interacting protein kinase 3 (RIPK3) on major adverse cardiovascular events (MACE) in patients with acute myocardial infarction (AMI) after percutaneous coronary intervention (PCI), as well as the predictive performance of RIPK3 combined with traditional cardiovascular risk factors.Methods:This study was a single-center prospective cohort study. It included patients with AMI who underwent PCI at Peking Union Medical College Hospital between September 2017 and November 2017. Baseline clinical data were collected, and plasma samples were obtained 6 hours after PCI to measure RIPK3 levels. Follow-up was conducted via outpatient visits or phone calls to record the occurrence of MACE, including cardiovascular death, hospitalization for heart failure, and vascular events (recurrent AMI or stroke). The predictive performance of RIPK3, traditional cardiovascular risk factors and their combination for MACE was compared using receiver operating characteristic (ROC) curves. Patients were divided into low- and high-RIPK3 level groups based on the optimal cutoff value of RIPK3. Multivariate Cox proportional hazards regression analysis was used to assess the impact of RIPK3 levels on MACE after PCI in AMI patients. Kaplan-Meier survival curves were plotted, and the log-rank test was used to compare MACE incidence between the low-and high-RIPK3 groups.Results:A total of 103 AMI patients who underwent PCI were included, aged 63.0 (56.0, 69.0) years, and 83 (80.6%) were male. The follow-up time was 5.17 (2.81, 5.17) years, during which 44 patients (42.7%) experienced MACE. The ROC curve analysis showed that the area under the curve ( AUC) for traditional cardiovascular risk factors was 0.68 (95% CI: 0.58-0.78), while the AUC for plasma RIPK3 was 0.72 (95% CI: 0.62-0.82). The combined AUC for traditional risk factors and RIPK3 was 0.75 (95% CI: 0.65-0.85). Multivariate Cox proportional hazards regression analysis indicated that plasma RIPK3 level is greater than or equal to the optimal cutoff value of 440.9 μg/L ( HR=3.31, 95% CI: 1.53-8.30, P=0.005) was an independent risk factor for MACE in AMI patients after PCI. Kaplan-Meier survival analysis demonstrated that the high-RIPK3 group had a significantly higher risk of MACE after PCI compared to the low-RIPK3 group (log-rank P=0.006). Conclusions:Elevated plasma RIPK3 level is an independent risk factor for MACE in AMI patients after PCI. Plasma RIPK3 combined with traditional cardiovascular risk factors can more effectively predict the occurrence of MACE in AMI patients after PCI. AMI patients with RIPK3≥440.9 μg/L have a higher risk of MACE after PCI.

Objective To analyze the relationship of total imaging burden score with blood pressure variability(BPV)and cognitive function in elderly patients with cerebral small vessel disease(CSVD).Methods Clinical data of 182 elderly CSVD patients admitted in our hospital from December 2022 to January 2024 were collected and analyzed retrospectively.According to the results of Chinese Guidelines for Diagnosis and Treatment of CSVD-Related Cognitive Dysfunc-tion(2019),Montreal Cognitive Assessment(MoCA)and other tools for their cognitive impair-ment,they were divided into cognitively impaired group(76 cases)and cognitively normal group(106 cases).The total burden score and BPV indicators[24 h systolic blood pressure coefficient of variation(24 h SBPCV),24 h diastolic blood pressure coefficient of variation(24 h DBPCV)]were compared between the two groups.ROC curve was plotted to evaluate the diagnostic value of total burden score,24 h SBPCV and 24 h DBPCV on cognitive impairment in elderly CSVD patients.Multivariate logistic regression analysis was used to identify the risk factors of cognitive impairment in the patients.Based on the MoCA score of the cognitively impaired group,these pa-tients were further assigned into mild and moderate-to-severe cognitive impairment subgroups[with a MoCA score of 18-25(43 cases)and＜18(33 cases),respectively].Then the total burden score,24 h SBPCV and 24 h DBPCV were compared between the two subgroups.Pearson correla-tion coefficient was adopted to evaluate the correlation of severity of cognitive impairment with total burden score,24 h SBPCV and 24 h DBPCV in the elderly CSVD patients.Results The total burden score,24 h SBPCV and 24 h DBPCV were significantly higher in the cognitively impaired group than the cognitively normal group(P＜0.01).ROC curve analysis indicated that total bur-den score(AUC=0.953,95％CI:0.926-0.980,P=0.000),24 h SBPCV(AUC=0.850,95％CI:0.795-0.906,P=0.000)and 24 h DBPCV(AUC=0.761,95％CI:0.690-0.832,P=0.000)had good diagnostic efficiency for cognitive impairment in the elderly CSVD patients,with a cut-off value of 1.5,11.82％,and 8.92％,respectively.Multivariate logistic regression analysis revealed that the above three indicators were risk factors for cognitive impairment in the elderly patients with CSVD(P＜0.05,P＜0.01).Their values were significantly lower in the mild than the moder-ate-to-severe cognitive impairment subgroups(P＜0.01).Pearson correlation analysis displayed that MoCA score was negatively correlated with total burden score,24 h SBPCV and 24 h DBPCV in elderly patients with CSVD(r=-0.755,-0.632,-0.601,P＜0.01).Conclusion Detection of total burden score and BPV indicators is beneficial to the assessment of cognitive impairment in elderly CSVD patients.The higher the total burden score and the greater the BPV indicators,the more severe the cognitive impairment is,which may explore new ideas for clinical diagnosis and treatment of CSVD.

Objective To analyze the application value of 3D Slicer software in endoscopic transsphenoidal surgery for pituitary adenoma resection.Methods A retrospective analysis was made on the clinical data of 36 patients with pituitary adenomas treated with 3D Slicer-assisted endoscopic transsphenoidal surgery(3D Slicer group)in the Department of Neurosurgery,The Second Affiliated Hospital of Xi'an Jiaotong University,from January 2024 to December 2024.Preoperatively,multimodal images were fused and reconstructed using 3D Slicer software to systematically evaluate bony anatomical structures such as sphenoid sinus ostia,intrasphenoidal septa,and sellar floor structures,design the size of pedicled nasoseptal flaps,and clarify the positional relationships between pituitary adenomas and surrounding vital structures including the internal carotid artery,pituitary gland,and optic chiasm,so as to provide real-time guidance for intraoperative procedures.Meanwhile,45 patients with pituitary adenomas treated with neuronavigation-assisted endoscopic transsphenoidal surgery from January 2023 to December 2023 were included as the control group(neuronavigation group).The surgical outcomes and postoperative complications were compared between the two groups.Results Compared with the neuronavigation group,the 3D Slicer group demonstrated higher identification rates of the optic nerve groove and carotid artery impression(94.4％vs.77.8％),shorter operative time[(2.9±0.6)h vs.(3.5±0.9)h],less intraoperative bleeding[(159.7±70.5)mL vs.(237.8±96.0)mL],and a lower incidence of postoperative olfactory dysfunction(8.3％vs.26.7％),with statistically significant differences(P＜0.05).However,there were no statistically significant differences between the two groups in the identification rate of the sphenoid sinus ostium(100.0％vs.97.8％),gross total resection rate(75.0％vs.64.4％),and the incidence of other postoperative complications,including cerebrospinal fluid leakage(0.0％vs.6.7％),intracranial infection(2.8％vs.11.1％),transient diabetes insipidus(30.6％vs.22.2％),and hypopituitarism(38.9％vs.37.8％,P＞0.05).Conclusion 3D Slicer software helps improve the mastery of anatomical basics in endoscopic transsphenoidal approach among junior and primary physicians,enhancing the clinical efficacy and safety of pituitary adenoma resection,and thus is worthy of clinical promotion.

Objective:To establish a risk prediction model by analyzing the influencing factors of delivery be-fore 34 weeks' gestatior with singleton pregnancies who underwent transcervical McDonald's cervical cerclage.Methods:Clinical data of 500 pregnant women with singleton pregnancies who underwent McDonald's cervical cerclage in the Department of Obstetrics and Gynecology,Heping Hospital,General Hospital of the People's Lib-eration Army Northern Theater of Operations from January 1,2019 to February 29,2024 were retrospectively ana-lyzed,and were randomly divided into a training set of 350 cases versus a validation set of 150 cases in accord-ance with a 7∶3 randomization.The data in the training set were screened for predictive variables affecting deliv-ery before 34 weeks' gestation after cervical cerclage using univariate combined multivariate Logistic regression,predictive modeling and assessment were established,and internal validation was performed in the validation set.Results:Preoperative cervical length(CL)＜20 mm,amniotic fluid deposits(AFS),absence of Lactobacillus vagi-nalis,and cervical dilatation＞3 cm were independent risk factors for adverse outcomes of labor and delivery be-fore 34 weeks' gestation after cervical cerclage(OR＞1,P＜0.05),and the Nomogram model constructed on the basis of this was found to have an AUC of 0.887,with a good agreement of the model(H-L test:P=0.905,Brier score of 0.095).The results of the clinical decision curve showed that the clinical decision value of the model was high when the threshold was between 0.10-0.92.The model was internally validated with an AUC of 0.812,and there was a high degree of agreement between the predictions and the actual situation(H-L test:P=0.859).Conclusions:The column-line diagram model constructed on this basis has a high clinical value,which can help us to screen out the high-risk pregnant women and give them individualized interventions in a timely manner.