Objective:To apply artificial intelligence（AI）in classifying ovarian tumors on ultrasound images，and compare the diagnostic results of several sonographers with varying seniority levels.Methods:A total of 645 patients diagnosed with adnexal masses via gynecological ultrasound examination at Qilu Hospital of Shandong University from January 2021 to December 2024 were enrolled. Three deep learning architectures，i.e.，Alexnet，Densenet121，and Resnet50 were developed and used to internally test the classification effectiveness of ovarian tumors，while the optimal model was selected for external testing. Two junior sonographers and two senior sonographers were recruited to independently diagnose ovarian tumors in the external test dataset. Subsequently，the benign and malignant results of the model's predictions were disclosed to each sonographer，and their revised diagnoses on the same external test data in combination with the best AI model were recorded.Results:The optimal model achieved an accuracy of 0.941，sensitivity of 0.936，and specificity of 0.944 on the internal test dataset，and maintained robust performance on the external test dataset with accuracy of 0.891，sensitivity of 0.880，and specificity of 0.907. Compared to junior sonographers，the optimal model demonstrated significantly higher sensitivity in discriminating benign from malignant ovarian tumors（0.880 vs. 0.723，0.602；all P<0.05）. No statistically significant difference was observed in diagnostic accuracy between the optimal model and senior sonographer 1（ P=0.05）. With assistance from the optimal model，junior sonographers achieved significant improvements in both sensitivity and specificity（sensitivity：0.723 vs. 0.843，0.602 vs. 0.819；specificity：0.778 vs. 0.833，0.685 vs. 0.741；all P<0.05）. Conclusions:The optimal model achieves comparable performance to that of senior sonographers in ovarian tumor classification. With model assistance，the diagnostic performance of junior sonographers is significantly improved.

Objective:To apply artificial intelligence（AI）in classifying ovarian tumors on ultrasound images，and compare the diagnostic results of several sonographers with varying seniority levels.Methods:A total of 645 patients diagnosed with adnexal masses via gynecological ultrasound examination at Qilu Hospital of Shandong University from January 2021 to December 2024 were enrolled. Three deep learning architectures，i.e.，Alexnet，Densenet121，and Resnet50 were developed and used to internally test the classification effectiveness of ovarian tumors，while the optimal model was selected for external testing. Two junior sonographers and two senior sonographers were recruited to independently diagnose ovarian tumors in the external test dataset. Subsequently，the benign and malignant results of the model's predictions were disclosed to each sonographer，and their revised diagnoses on the same external test data in combination with the best AI model were recorded.Results:The optimal model achieved an accuracy of 0.941，sensitivity of 0.936，and specificity of 0.944 on the internal test dataset，and maintained robust performance on the external test dataset with accuracy of 0.891，sensitivity of 0.880，and specificity of 0.907. Compared to junior sonographers，the optimal model demonstrated significantly higher sensitivity in discriminating benign from malignant ovarian tumors（0.880 vs. 0.723，0.602；all P<0.05）. No statistically significant difference was observed in diagnostic accuracy between the optimal model and senior sonographer 1（ P=0.05）. With assistance from the optimal model，junior sonographers achieved significant improvements in both sensitivity and specificity（sensitivity：0.723 vs. 0.843，0.602 vs. 0.819；specificity：0.778 vs. 0.833，0.685 vs. 0.741；all P<0.05）. Conclusions:The optimal model achieves comparable performance to that of senior sonographers in ovarian tumor classification. With model assistance，the diagnostic performance of junior sonographers is significantly improved.

Objective To evaluate the efficacy of different treatment options for cholecysto-choledocholithiasis (CCL),and try to find the ideal treatment.Methods From January 2006 to January 2016,a total of 3 107 patients with CCL from the Affiliated Hospital of Inner Mongolia Medical University were enrolled.Among them,1 283 patients were in open cholecystectomy (OC) and open common bile duct exploration (OCBDE) group,964 patients were in laparoscopic cholecystectomy (LC) and laparoscopic common bile duct exploration (LCBDE) group,and 860 patients were in endoscopic retrograde cholangiopancreatography (ERCP) + LC group.The clinical data of the three groups were analysed.One-way analysis of variance and chi-square test were performed for statistical analysis.Results From 2006 to 2010,the percentage of patients treated with OC + OCBDE,LC + LCBDE,and ERCP + LC were 56.05％ (829/1 479),25.15％ (372/1 479) and 18.80％ (278/1 479),respectively;from 2011 to 2016,the percentage of patients received the above three treatments were 27.89％ (454/1 628),36.36％ (592/1 628) and 35.75％ (582/1 628),respectively.The difference in the proportion of the same treatment at different times was statistically significant (x2 =4.775,4.168 and 0.669,all P ＜ 0.05).The success rate of surgery in the OC + OCBDE group was 100.00％ (1 370/1 370);while the success rate of surgery in the LC + LCBDE group was 94.26％ (920/976),and 56 patients converted to OC + OCBDE;the success rate of surgery in the ERCP + LC group was 95.00％ (817/860),and 31 patients converted to OC + OCBDE,and 12 patients to LC + LCBDE.The intraoperative complication in OC + OCBDE,LC + LCBDE and ERCP + LC were 2.85％ (39/1 370),3.48％ (32/920) and 1.22％ (10/817),respectively.The incidence rates of postoperative complication were 4.89％ (67/1 370),5.34％ (50/920) and 5.51％ (45/817),respectively.The incidence rates of intraoperative complication of the ERCP + LC group was lower than that of OC + OCBDE group and LC + LCBDE group,and the differences were statistically significant (x2 =6.203 and 3.001;both P ＜ 0.05).However there was no significant difference in incidence rate of postoperative complications among the three groups (all P ＞ 0.05).The hospital stay of the OC + OCBDE group,the LC + LCBDE group and the ERCP + LC group were (6.7 ± 1.3) days,(5.6 ± 1.2) days and (10.9 ± 1.6) days,respectively,and the differences were statistically significant (F =90.010,P ＜ 0.01).The hospitalization expenses of OC + OCBDE group,LC + LCBDE group and ERCP+LC group were (13 720±1 910) yuan,(18 150±1 490) yuan and (25 830 ± 2 430) yuan,respectively,and the differences were statistically significant (F =302.991,P ＜ 0.01).Conclusion The first choice of patients with CCL is endoscopic minimally invasive treatment and open surgery can be used as a remedial method for endoscopic treatment.

3D printing service in medical librariescan help their users to master the new medical technologies, meet the needs in clinical teaching, and expand their value-added service.The application of 3D printing techno logy in 5 American medical libraries was thus analyzed in terms of its equipments, establishment of biomedical 3D model, and charge of 3D printing service, with the problems stressed for domestic medical libraries in providing 3D printing service, such as rational allocation of fees, solution of intellectual properties, and training of professional librarians.

Objective To describe body mass index (BMI) changes in migrant children with overweight and obesity in Beijing treated by the exercise and dietary intervention programme.Methods BMI was assessed at before and after the intervention in 30 migrant obesity and overweight children aged 9 ～ 12 in two schools from March,2010.Results BMI reduced significantly one week after the intervention ( P ＜ 0.05 ),but from two months to six months after the intervention BMI increased again to the pretreatment level( P ＜ 0.05 ).Conclusion The intervention was successful in decreasing BMI,however maintenance of BMI after treatment was not easy.The controlling of obesity and overweight is a long term process which needs the cooperation and supervision of their parents.