OBJECTIVE: To compare the performance of a photostimulable phosphor (PSP) imaging system with radiographic film for the detection of approximal lesions under two background lighting conditions. METHODS: Bitewing exposures were made of the extracted human teeth with artificial approximal lesions using D-speed and E-speed radiographic film and a PSP imaging plate. Eight dentists rated 12 approximal surfaces in each image on a 5-point scale. Images were viewed with overhead fluorescent room lighting on and off. ROC curves were generated for each image receptor and lighting condition. Repeated measures analysis of variance ANOVA tested differences between observers, lesion size, image receptor, lighting condition. RESULTS: Mean ROC curve areas for D-speed film viewed in full ambient light scored highest (0.8030) followed by E-speed film using dark conditions (0.7386). The lowest mean curve area was found when PSP (enhanced mode) images were viewed in a dark room (0.6726). Significant differences were found between observers, lesion size, image receptor and image receptor-lesion size interaction. No significant difference was found with background lighting. CONCLUSIONS: Background lighting does not appear to effect the ability to detect artificial approximal lesions.
OBJECTIVE: To compare the performance of a photostimulable phosphor (PSP) imaging system with radiographic film for the detection of approximal lesions under two background lighting conditions. METHODS: Bitewing exposures were made of the extracted human teeth with artificial approximal lesions using D-speed and E-speed radiographic film and a PSP imaging plate. Eight dentists rated 12 approximal surfaces in each image on a 5-point scale. Images were viewed with overhead fluorescent room lighting on and off. ROC curves were generated for each image receptor and lighting condition. Repeated measures analysis of variance ANOVA tested differences between observers, lesion size, image receptor, lighting condition. RESULTS: Mean ROC curve areas for D-speed film viewed in full ambient light scored highest (0.8030) followed by E-speed film using dark conditions (0.7386). The lowest mean curve area was found when PSP (enhanced mode) images were viewed in a dark room (0.6726). Significant differences were found between observers, lesion size, image receptor and image receptor-lesion size interaction. No significant difference was found with background lighting. CONCLUSIONS: Background lighting does not appear to effect the ability to detect artificial approximal lesions.