Suppr超能文献

通过计算机超声图像数字化改善肾移植集合系统扩张的评估。

Improving the evaluation of renal transplant collecting system dilatation by computerized ultrasound image digitization.

作者信息

Kashi S H, Romaniuk C, Sadek S A, Irving H C

机构信息

University Department of Surgery, St James's University Hospital, Leeds, UK.

出版信息

Br J Radiol. 1993 Nov;66(791):1002-8. doi: 10.1259/0007-1285-66-791-1002.

Abstract

43% of the renal allografts transplanted at our centre demonstrate pelvicalyceal dilatation on ultrasonography. However, only 6% of grafts are ultimately shown to be obstructed by antegrade pyelography. We have attempted to improve the specificity of ultrasonography by a quantitative analysis of 394 serial ultrasound studies from 150 renal transplant recipients. Representative images in the longitudinal and transverse axes were selected by a single observer for each of the studies and the outline of the kidney and its pelvicalyceal system was traced out. The images were then digitized using a Summagraphics tablet so that the length, width and area of the kidney and its pelvicalyceal system were calculated for each image. A step-wise discriminant analysis was performed which showed that the ratio of the area of the kidney to its pelvicalyceal area in both axes discriminated between normal, dilated but non-obstructed, and obstructed grafts. The median ratio of kidney to collecting system area in the obstructed grafts was 5.2 in the longitudinal and 4.4 in the transverse scan planes (vs median ratios of 32.8 and 22.0 for non-dilated grafts). Using canonical discriminant function, an index was derived which combines these ratios for the longitudinal and transverse images of the same kidney and this has further improved the diagnostic accuracy (median index of 2.45 in obstructed group vs -0.85 in non-dilated kidneys). This method improves upon the radiologist's report and identifies those grafts which require intervention.

摘要

在我们中心移植的肾移植受者中,43%的移植肾在超声检查中显示肾盂肾盏扩张。然而,只有6%的移植肾最终经顺行肾盂造影显示存在梗阻。我们试图通过对150名肾移植受者的394次连续超声研究进行定量分析来提高超声检查的特异性。由一名观察者为每项研究选择纵轴和横轴的代表性图像,并勾勒出肾脏及其肾盂肾盏系统的轮廓。然后使用Summagraphics数位板将图像数字化,以便计算每个图像中肾脏及其肾盂肾盏系统的长度、宽度和面积。进行了逐步判别分析,结果表明,两个轴上肾脏面积与其肾盂肾盏面积的比值可区分正常、扩张但无梗阻以及梗阻的移植肾。梗阻移植肾在纵轴扫描平面上肾脏与集合系统面积的中位数比值为5.2,横轴扫描平面上为4.4(非扩张移植肾的中位数比值分别为32.8和22.0)。使用典型判别函数,得出了一个指数,该指数结合了同一肾脏纵轴和横轴图像的这些比值,进一步提高了诊断准确性(梗阻组的中位数指数为2.45,非扩张肾脏为-0.85)。这种方法改进了放射科医生的报告,并识别出那些需要干预的移植肾。

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验