Diagnostic Accuracy of Digitized Images Using Different Resolution Settings of Digital Camera in Detection of Proximal Caries

Objective: When none of digital systems and scanners is accessible and it is essential to have digitized images of conventional radiographs, digital cameras can be used. The Aim of this study was to investigate whether digital images obtained by different resolutions of a digital camera are matched to the original radiographs in evaluation of caries. Methods: In this diagnostic accuracy in vitro study the conventional radiographs of168 proximal surfaces of 84 teeth were produced, Then they were digitized with digital camera in three different resolutions high (2048x1536), medium (1600x1200) and low resolution (480x460). Images were stored in Photoshop7.0 software, and were evaluated by5 observers to show the presence and depth of the caries. Cronbach’s &alpha calculated inter-observers agreement and in order to calculate the agreement with original conventional radiographs Kappa index was used. Results: In assessing the presence of caries, the agreement between low, medium and high resolutions with original radiographs were 0.286, 0.235 and 0 respectively. Also, assessing the depth of the caries agreement was reported0.21, 0.338 and 0.412 respectively. In most instances, there was a fair agreement between the different resolutions and original radiographs. The highest inter-observer’s agreement was reported in diagnosis of the presence of the caries with using high resolution (&alpha=0.837) and the lowest inter-observer’s agreement was reported in diagnosis of the depth of the caries with medium resolution(&alpha=0.762).There was no significant difference reported in observations of different resolutions and original images. Conclusion: Using of high-resolution cameras did not show a significant difference with medium and low resolutions in caries evaluations. Therefore, considering the increase in the file size and difficulties in cameras selection, using of high-resolution digital cameras is not necessary in order to increase the diagnostic accuracy of digitized images.