Yoshino M T, Carmody R, Fajardo L L, Seeger J, Jones K
Department of Radiology, University of Arizona, Tucson 85724.
Invest Radiol. 1992 Jan;27(1):55-9. doi: 10.1097/00004424-199201000-00011.
This study was done to assess the diagnostic accuracy of high-resolution (5 lp/mm) teleradiology for detecting cervical spine fractures. Single radiographs from 25 patients with and 25 patients without cervical spine fractures were transmitted between two units of our teleradiology system (Dupont DTR 2000) located 5 miles apart. Each image was examined by four readers. Fracture detection accuracy was assessed by generating receiver operating characteristic (ROC) curves and comparing the areas under each reader's curves for original and transmitted images. Two readers had statistically significant better fracture detection using nontransmitted images, whereas two had no significant differences in accuracy. The authors conclude that high resolution in and of itself is not adequate for fracture detection, and that issues concerning image contrast manipulation also will have to be addressed before teleradiology systems can be used for clinical cervical spine fracture screening.
本研究旨在评估高分辨率（5线对/毫米）远程放射学检测颈椎骨折的诊断准确性。在相距5英里的我们的远程放射学系统（杜邦DTR 2000）的两个单元之间传输了25例有颈椎骨折患者和25例无颈椎骨折患者的单张X线片。每位阅片者检查每幅图像。通过绘制受试者工作特征（ROC）曲线并比较每位阅片者对原始图像和传输图像的曲线下面积来评估骨折检测准确性。两名阅片者使用未传输图像时骨折检测在统计学上有显著更好的表现，而另外两名阅片者在准确性上无显著差异。作者得出结论，高分辨率本身不足以进行骨折检测，并且在远程放射学系统可用于临床颈椎骨折筛查之前，还必须解决有关图像对比度处理的问题。
