Observers' use of image enhancement in assessing caries in radiographs taken by four intra-oral digital systems

Objectives. To evaluate observers' use of image-enhancement facilities and time consumption in assessing caries in radiographs taken with four digital systems. Methods. In total, 131 extracted human premolars and molars were mounted three in a line. Radiographs were taken using four digital systems: Digora (DIG), Radio VisioGraphy (RVG), Sens-A-Ray (SAR) and Visualix (VIX), and imported into a programme with routines for adjustment of brightness, contrast and gamma curve. Sixteen images from each digital system were compressed (JPEG, irreversible compression). The 588 images were scored by six observers for approximal and occlusal caries on a five-point confidence scale using enhancement as they pleased. The programme automatically recorded any enhancement made without the observers knowing this. Results. Some form of digital enhancement was used in almost all images, with the gamma curve being the most frequent. The VIX images were enhanced most followed by SAR, DIG and RVG images. The differences were significant (p < 0.01) except between DIG and SAR images. The compressed images were enhanced significantly more than their uncompressed counterparts (p = 0.02). The average time spent recording one image was 24 s. On average, significantly less time was spent with the DIG images than the other systems (p < 0.01), while there were no significant differences between the CCD-based systems (p > 0.2). There was no relationship between time spent and number of manipulations performed. Conclusions. The observers took advantage of the facilities available for enhancement of density and contrast in digital images. The potential of gamma curve manipulation requires further investigation.