Objective:To develop and evaluate a medical visual question answering (VQA) model for neonatal lung ultrasound (LUS) images to enhance intelligent auxiliary diagnosis of neonatal pulmonary diseases.Methods:Using data from neonates admitted to Beijing Obstetrics and Gynecology Hospital, Capital Medical University (January 2023 to December 2024), an image-question-answer dataset comprising 251 LUS images was constructed [43 pneumonia (17.1%), 42 neonatal respiratory distress syndrome (16.7%), 83 transient tachypnea (33.1%), and 83 normal (33.1%) images] with a four-tier medical question-answer framework. Building upon the Qwen2.5-VL-7B base model and integrating LoRA fine-tuning with chain-of-thought prompting, we developed the NLUS-VQA model to enhance visual-language semantic alignment and enable stepwise clinical reasoning, achieving efficient small-sample adaptation. Model performance was comprehensively assessed through natural language generation metrics (BLEU-4, ROUGE-1/2/L), qualitative evaluation of characteristic recognition, and clinical consistency analysis.Results:(1) Quantitative evaluation demonstrated that NLUS-VQA achieved scores of 22.38 (BLEU-4), 48.26 (ROUGE-1), 22.40 (ROUGE-2), and 37.20 (ROUGE-L), representing significant improvements over baseline models. (2) Qualitatively, the model exhibited strong performance in identifying lung consolidation, coalescent B-lines, and snowflake signs, with its chain-of-thought strategy enhancing clinical interpretability and answer accuracy. (3) Clinically, NLUS-VQA achieved a Cohen's Kappa coefficient of 0.78 and diagnostic accuracy of 80.8% (21/26), indicating substantial agreement with clinical experts.Conclusion:The NLUS-VQA model demonstrates robust interpretability in recognizing key sonographic patterns (e.g. lung consolidation, confluent B-lines, and snowflake signs), providing a scalable framework for small-sample medical image analysis, though diagnostic performance on complex conditions remains limited by dataset scale and minority class representation.

Objective:To develop and evaluate a medical visual question answering (VQA) model for neonatal lung ultrasound (LUS) images to enhance intelligent auxiliary diagnosis of neonatal pulmonary diseases.Methods:Using data from neonates admitted to Beijing Obstetrics and Gynecology Hospital, Capital Medical University (January 2023 to December 2024), an image-question-answer dataset comprising 251 LUS images was constructed [43 pneumonia (17.1%), 42 neonatal respiratory distress syndrome (16.7%), 83 transient tachypnea (33.1%), and 83 normal (33.1%) images] with a four-tier medical question-answer framework. Building upon the Qwen2.5-VL-7B base model and integrating LoRA fine-tuning with chain-of-thought prompting, we developed the NLUS-VQA model to enhance visual-language semantic alignment and enable stepwise clinical reasoning, achieving efficient small-sample adaptation. Model performance was comprehensively assessed through natural language generation metrics (BLEU-4, ROUGE-1/2/L), qualitative evaluation of characteristic recognition, and clinical consistency analysis.Results:(1) Quantitative evaluation demonstrated that NLUS-VQA achieved scores of 22.38 (BLEU-4), 48.26 (ROUGE-1), 22.40 (ROUGE-2), and 37.20 (ROUGE-L), representing significant improvements over baseline models. (2) Qualitatively, the model exhibited strong performance in identifying lung consolidation, coalescent B-lines, and snowflake signs, with its chain-of-thought strategy enhancing clinical interpretability and answer accuracy. (3) Clinically, NLUS-VQA achieved a Cohen's Kappa coefficient of 0.78 and diagnostic accuracy of 80.8% (21/26), indicating substantial agreement with clinical experts.Conclusion:The NLUS-VQA model demonstrates robust interpretability in recognizing key sonographic patterns (e.g. lung consolidation, confluent B-lines, and snowflake signs), providing a scalable framework for small-sample medical image analysis, though diagnostic performance on complex conditions remains limited by dataset scale and minority class representation.

Objective To evaluate the efficiency and safety of drug-eluting stents (DES) and bare metal stents (BMS) in elderly with coronary artery disease (CAD).Methods PubMed,Embase,Cochrane Library,CNKI and Wanfang databases were searched before October 28,2017.The pertinent randomized controlled trials (RCTs) and cohort studies (CS) were included.The outcomes were all-cause mortality (ACM),myocardial infarction (MI),target vessel revascularization (TVR),cardiac death (CD),stent thrombosis (ST) and bleeding.Relative risk (RR) with 95％ confidence interval (95％CI) was analyzed.Results A total of 17 studies with 10 813 patients were included.Overall and in CS group,there were significant lower risks of ACM (RR＝0.71,0.68,both P＜0.001),MI (RR＝0.66,P＜0.001;RR＝0.60,P＝0.008),TVR (RR＝0.50,0.54,both P＜0.001),CD (RR＝0.73,0.72,P＝0.003,0.020) and ST (RR＝0.68,0.59,P＝ 0.02,0.01) in DES patients compared with BMS patients,but the risk of bleeding was not significantly different (RR ＝ 1.00,1.00,P ＝ 0.96,0.97).In RCT group,DES (compared with BMS) was only associated with significantly lower risks of MI (RR＝0.68,P＝0.003) and TVR (RR＝0.43,P＜0.001),no difference was detected in other indexes.Conclusion DES is more efficacious and safer than BMS for elderly patients with CAD.

Objective To obtain the normal reference ranges of global longitudinal strain (GLS),global circumferential strain (GCS),global area strain (GAS) and global radial strain (GRS) of left ventricular in normal adults by three-dimensional speckle tracking imaging (3D-STI) using Meta analysis.Methods Eligible trials which detected global strain of left ventricular in normal subject through 3D-STI were searched in Embase,Pubmed,Cochrane Library database.According to the heterogeneity,parameters of contained studies were analyzed the weighted mean difference (WMD) and 95％ confidence interval (CI).The statistical software was STATA 12.0.Results Totally 1 552 healthy adults from 27 articles were included.Based on the Meta-analysis,theWMDand 95％CIofGLSwere 17.80 and (16.27,19.33),of GCS were 24.73 and (22.50,26.95),of GRS were 47.86 and (39.52,56.19),of GAS were 36.17 and (34.08,38.26).Conclusion The Meta analysis defines reference range of strains obtained by 3D-STI in healthy adults.Using these parameters of 3D globe strains,a guidance of reference for patienfs management and therapy selection may be provided.