EVALUATION OF MONITOR PERFORMANCE IN CARIES DETECTION

Background Digital imaging has become a staple of dentistry for diagnosis and treatment planning. Off-the-shelf displays are most commonly used to view acquired images in bright light compared with medical radiology, which almost exclusively uses autocalibrating high-resolution medical grade displays. Objective(s) This study aimed to assess whether medical-grade displays perform better in the detection of confirmed caries on extracted teeth compared with off-the-shelf monitors. The null hypothesis stated that there is no significant difference between medical-grade and off-the-shelf displays. Study Design Thirty noncarious premolars and 31 carious premolars that were extracted for therapeutic reasons were examined independently by 2 evaluators, who visually (with magnification) verified the presence/absence of caries. Teeth were placed in the second premolar area of a model with 2 noncarious molars and a noncarious premolar with standardized distance and positioning from the x-ray source. Teeth were imaged with a Dexis (DEXIS. LLC, Hatfield, PA) sensor using standardized exposure parameters (7 mA, 0.08 seconds, 65 kVp). Images were randomized. Two independent, calibrated, and masked evaluators scored the images on a 5-point Likert scale for the presence or absence of caries on 4 different monitors under standardized ambient light conditions: BARCO 3 MP medical grade, a prototype BARCO 2 MP, DELL Ultrasharp, and WIDE 5 MP monochrome. Results Barco 3 MP and Wide 5 MP monitors led to greater accuracy compared with Barco 2 MP and Dell Ultrasharp monitors with a difference of at least 10%. Barco 3 MP was most sensitive, followed by Barco 2 MP and Wide 5 MP. Specificity was highest for Barco 3 MP and Dell, followed by Barco 2 MP and Wide 5 MP. The interobserver agreement was highest for the Barco 2 MP and Dell displays. None of the findings was statistically significant. Discussion/Conclusions Although there seems to be more information available on higher-resolution monitors with better accuracy in identifying carious lesions, these differences are not statistically significant.