Background Chronic kidney disease（CKD） patients are at high risk of atherosclerosis. Structural and elastic changes of carotid artery wall reflect the range and degree of atherosclerosis in peripheral arteries, whic...Background Chronic kidney disease（CKD） patients are at high risk of atherosclerosis. Structural and elastic changes of carotid artery wall reflect the range and degree of atherosclerosis in peripheral arteries, which can be acquired by ultrasound radiofrequency-data technique automatically and precisely. Methods A total of 66 CKD patients with negative results on routine carotid artery ultrasound examination were enrolled, and 30 healthy physical examinees were selected as controls. Patients were divided into 3 groups according to CKD stage： stage 1-2,stage 3-4 and stage 5. Clinical characteristics and the laboratory results were acquired. Intima-media thickness（IMT） and compliance coefficient（CC） of common carotid artery were measured by ultrasound radiofrequencydata technique（QIMT and QAS）. Predictors of IMT and CC were analyzed respectively. Results Among the 66 patients,15 were on stage 1-2, 15 on stage 3-4 and 36 on stage 5 according to e GFR. The common carotid artery IMT（CCIMT）of all the CKD groups except patients on stage 1-2 was significantly increased when compared with controls. The CC of carotid artery significantly was decreased in every CKD group compared with controls. Age and CKD stage were significant predictors of CCIMT and CC in CKD patients（P〈0.05）. Aging and advanced CKD stage were associated with increased CCIMT（OR=4.855 and 4.969） and decreased CC（OR=32.178 and 14.068）. Conclusions Radiofrequency-data technique can detect the small changes of structure and elasticity of carotid artery wall in CKD patients. CKD patients have increased IMT and decreased elasticity of carotid artery compared with healthy subjects. Aging and advanced CKD stage are associated with increased CCIMT and decreased CC.展开更多
目的:对比分析不同年资超声科医师应用人工智能(artificial intelligence,AI)S-Detect技术诊断乳腺结节的诊断效能,探讨S-Detect技术的临床应用价值。方法:对2020年7—10月在上海市长宁区妇幼保健院行乳腺结节手术的100例患者(144例结...目的:对比分析不同年资超声科医师应用人工智能(artificial intelligence,AI)S-Detect技术诊断乳腺结节的诊断效能,探讨S-Detect技术的临床应用价值。方法:对2020年7—10月在上海市长宁区妇幼保健院行乳腺结节手术的100例患者(144例结节),分别由1名高年资副主任医师和1名低年资主治医师行常规临床超声诊断,按照乳腺影像报告和数据系统(Breast Imaging Reporting and Data System,BI-RADS)分类标准,以分类结果≤4A诊断为良性结节,>4A诊断为恶性结节;并分别应用S-Detect技术评价结节,以纵切及横切两次结果均为“可能良性”诊断为良性结节,否则诊断为恶性结节。以病理学检查结果为金标准,对比分析常规超声检查及S-Detect技术2名医师的诊断效能及一致性。结果:144个乳腺结节中,良性结节124个,恶性结节20个。使用常规超声检查高年资医师及低年资医师的受试者工作特征(receiver operating characteristic,ROC)曲线的曲线下面积(area under curve,AUC)分别为0.868和0.690;S-Detect技术高年资医师及低年资医师的AUC分别为0.877和0.893;常规超声检查低年资医师的AUC与其他三者差异有统计学意义(P<0.01)。常规超声检查高年资医师及低年资医师组内相关系数(intraclass correlation coefficient,ICC)为0.736,具有中度一致性;S-Detect技术高年资医师及低年资医师ICC为0.928,具有高度一致性。结论:临床不同年限超声科医师工作经验对于乳腺结节的鉴别诊断有影响,但是AI技术可以降低经验对诊断结果的影响。S-Detect技术值得临床推广。展开更多
基金supported by Medical Scientific Research Foundation of Guangdong Province(No.201611216254466)
文摘Background Chronic kidney disease（CKD） patients are at high risk of atherosclerosis. Structural and elastic changes of carotid artery wall reflect the range and degree of atherosclerosis in peripheral arteries, which can be acquired by ultrasound radiofrequency-data technique automatically and precisely. Methods A total of 66 CKD patients with negative results on routine carotid artery ultrasound examination were enrolled, and 30 healthy physical examinees were selected as controls. Patients were divided into 3 groups according to CKD stage： stage 1-2,stage 3-4 and stage 5. Clinical characteristics and the laboratory results were acquired. Intima-media thickness（IMT） and compliance coefficient（CC） of common carotid artery were measured by ultrasound radiofrequencydata technique（QIMT and QAS）. Predictors of IMT and CC were analyzed respectively. Results Among the 66 patients,15 were on stage 1-2, 15 on stage 3-4 and 36 on stage 5 according to e GFR. The common carotid artery IMT（CCIMT）of all the CKD groups except patients on stage 1-2 was significantly increased when compared with controls. The CC of carotid artery significantly was decreased in every CKD group compared with controls. Age and CKD stage were significant predictors of CCIMT and CC in CKD patients（P〈0.05）. Aging and advanced CKD stage were associated with increased CCIMT（OR=4.855 and 4.969） and decreased CC（OR=32.178 and 14.068）. Conclusions Radiofrequency-data technique can detect the small changes of structure and elasticity of carotid artery wall in CKD patients. CKD patients have increased IMT and decreased elasticity of carotid artery compared with healthy subjects. Aging and advanced CKD stage are associated with increased CCIMT and decreased CC.
文摘目的:对比分析不同年资超声科医师应用人工智能(artificial intelligence,AI)S-Detect技术诊断乳腺结节的诊断效能,探讨S-Detect技术的临床应用价值。方法:对2020年7—10月在上海市长宁区妇幼保健院行乳腺结节手术的100例患者(144例结节),分别由1名高年资副主任医师和1名低年资主治医师行常规临床超声诊断,按照乳腺影像报告和数据系统(Breast Imaging Reporting and Data System,BI-RADS)分类标准,以分类结果≤4A诊断为良性结节,>4A诊断为恶性结节;并分别应用S-Detect技术评价结节,以纵切及横切两次结果均为“可能良性”诊断为良性结节,否则诊断为恶性结节。以病理学检查结果为金标准,对比分析常规超声检查及S-Detect技术2名医师的诊断效能及一致性。结果:144个乳腺结节中,良性结节124个,恶性结节20个。使用常规超声检查高年资医师及低年资医师的受试者工作特征(receiver operating characteristic,ROC)曲线的曲线下面积(area under curve,AUC)分别为0.868和0.690;S-Detect技术高年资医师及低年资医师的AUC分别为0.877和0.893;常规超声检查低年资医师的AUC与其他三者差异有统计学意义(P<0.01)。常规超声检查高年资医师及低年资医师组内相关系数(intraclass correlation coefficient,ICC)为0.736,具有中度一致性;S-Detect技术高年资医师及低年资医师ICC为0.928,具有高度一致性。结论:临床不同年限超声科医师工作经验对于乳腺结节的鉴别诊断有影响,但是AI技术可以降低经验对诊断结果的影响。S-Detect技术值得临床推广。
