Abstract
Objective: Routine direct radiograms of the front chest (chest X-ray radiograms) are normally interpreted by displaying 2 radiograms side-by-side and comparing them (conventional method). However, in the filmless environment in our health and preventive medicine center, we mainly conduct comparative examinations of X-ray radiograms using afterimages produced by paging using rapid wheel scrolling on radiology workstations (afterimage method). A comparative study of the 2 methods demonstrated the utility of the afterimage method.
Methods: In January and February 2011, 20 doctors conducted a comparison of interpreting the chest X-ray radiograms of 20 patients by the conventional method and afterimage method, using a NV-1000 radiology workstation.
Results: At 0.945, the sensitivity of the afterimage method was 1.5 times better than the conventional method. The likelihood ratio for a positive finding was 15.8, 1.8 times that of the conventional method and the likelihood ratio for a negative finding was 0.06, 0.14 times that of the conventional method. The false negative rate of the afterimage method decreased by around 85%. There was hardly any difference between the 2 methods in terms of specificity and interpretation time. The afterimage method was significantly superior with regard to ROC analysis by the pool method.
Conclusion: A comparison of interpreting the X-ray radiograms by the conventional and afterimage methods demonstrated the clear superiority of the latter. While there was a slight problem with reproducibility when taking radiograms, this should be able to be overcome through refinement of the method of taking radiograms and enhancement of radiology workstation functions, which should further raise accuracy. In the future, its is expected that this technique will be widely adopted in health check-ups as well as in the clinical setting, where it should contribute to raising accuracy in the early discovery of lung cancer and a wide range of other diseases, as well as in patient follow-up.
