Minimum Resolution for Caries Diagnosis Using Digitized Images

Objective: Questions exist about what is the required scanning resolution needed when using computer monitors to evaluate radiographs for caries diagnosis. The higher the resolution the larger is the file size and larger are the storage requirements. Thus, the purpose of this study was to evaluate scanned bitewing radiographs of artificial carious lesions in the enamel of extracted caries-free teeth. Methods: Artificial carious lesions in enamel were created on 27 proximal surfaces of 40 caries-free teeth that were then mounted in blocks in groups of eight. Each block was imaged using Insight Film (Kodak, Rochester, NY. Film images were randomly selected to represent 1 for each block and these were digitized using an Epson Expression 1600 flatbed scanner (Epson Electronics America, Inc., San Jose, CA). Images were scanned at 100, 200, 400, 800 and 1600 ppi and saved as uncompressed TIFFs for later viewing. These resolutions had been previously found to be equivalent to 2.9, 4.5, 6.7 and 7.7 lp/mm (1600 not used). Radiographs were viewed in a clinical setting in mixed fluorescent daylight and natural daylight. A 17-inch Optiquest Z70 monitor (Viewsonic, Walnut, CA) running Photoshop 5.5 (Adobe Systems Inc., San Jose, CA) for Windows 98 (Microsoft, Redmond, WA) operating system was used. Images were coded, randomly sequenced, adjusted with the fit on screen command and brightness-contrast adjustments made by each of six examiners. Proximal surfaces in contact were scored using a 5-point scale. Data were evaluated using receiver operating characteristics analysis (ROCFIT, CE Metz, Chicago, IL). Results: Results failed to show a significant difference in the ability to observe carious lesions images that ranged from 4.5 to 7.7 lp/mm. Conclusion: These results suggest that the minimum level of resolution is between 2.9 (100 ppi) and 4.5 (200 ppi) lp/mm and that higher resolution is not necessary.