Purpose: To define the value of teleradiographic studies, a comparison was carried out between digitised copies of CT examinations of the skull with the original images. Differences in image quality obtained from a digital scanner and a camera were quantified.
Material and method: 56 CT examinations of the skull, 28 of which had discrete abnormalities, were chosen for ROC analysis. The original films were digitised with a Vidar VXR-12 scanner and Panasonic WV-160 and WV-BP 500 cameras. The images were evaluated by five radiologists after image transfer with Video Conference software to a personal computer.
Results: For the analysis of the films the area under the ROC curve was 0.91 +/- 0.04, for the digital scanner it was 0.85 +/- 0.04, for camera WV-BP 500 0.89 +/- 0.06 and for camera WE-160 0.87 +/- 0.09. Comparison with the film findings showed. a minimal p-value of 0.17 which indicated that there was no significant reduction in diagnostic value following digitization.
Conclusion: The probable reason for the slight deterioration using the digital scanner was the reduction to 75 dpi compared with 134 dpi on the CT films. The cameras produce image noise comparable to CT with low window settings and reduced local resolution. We expect similar results for CT with soft tissue windows or for MRT of the skull. Conventional radiographs containing high local resolution, wide grey scale and low image noise would presumably make higher demands on methods of digitization.