Stassinakis A, Zeyer O, Brägger U
Klinik für Zahnerhaltung, Universität Bern.
Schweiz Monatsschr Zahnmed. 1995;105(12):1539-45.
The purpose of this study was to evaluate the possibility to delineate artificial osseous lesions with direct digital radiography (DDR) in comparison to conventional radiography (Rx). Defined lesions of nine different depths were created on ten mandibular segments of pigs. A conventional and a direct digital radiography was made of each lesion. 400 pairs of images were evaluated by ten dentists on the basis of a 5-point rating scale. The statistical evaluation included the interexaminer agreement and ROC-analyses. The mean interexaminer agreement was low and not significantly different between the two methods (Rx: kappa = 21.87 +/- 5.08; DDR: kappa = 20.25 +/- 3.79; p > 0.05). The ROC-analyses showed no significant differences between the two methods for the smallest lesions (Rx: 0.72 +/- 0.12; DDR: 0.64 +/- 0.12; p > 0.05). Defects between 0.5 mm and 1.0 mm of size were significantly better depicted with conventional radiographs (p > 0.05). Lesions from 1.25 to 2.0 mm were statistically not significantly different (p > 0.05). The diagnosis of lesions of 1.0 mm diameter and less than 1.25 mm depth was better delineated by conventional radiographs than by direct digital radiographs. However, proper diagnostic results were only achieved in lesions bigger than 2.0 mm. During the time this study was conducted many technical improvements were introduced for direct digital equipment with higher spatial resolutions. The presented methodology is proposed for the examination of the diagnostic properties of new imaging tools in comparison to conventional radiography.