Interaction between noise and file compression and its effect on the recognition of caries in digital imaging

Objectives: To determine the interaction between image noise and file compr ession, with special emphasis on the accuracy of caries diagnosis. Methods: Fifty-nine bitewing radiographs of patients were taken simultaneou sly with Ektaspeed Plus (Eastman-Kodak, Rochester, NY, USA) film without le ad foil and the Digora(R) storage phosphor system (Soredex, Helsinki, Finla nd). Three different levels of Gaussian noise were added to the original di gital images which were then compressed with JPEG 53. Seven observers evalu ated the presence and depth of caries lesions on selected approximal surfac es on a 5-point scale. The results of JPEG 27 compression from a previous s tudy were also included. ROC analysis was used together with multivariate a nalysis of variance (MANOVA). Results: JPEG 27 and 53 reduced the file size down to 7% and 4.6% of the or iginal respectively. ROC curve analysis showed no significant difference be tween image conditions (original, JPEG 27, and JPEG 53) at the same noise l evel. JPEG 27 and 53 had larger A, scores than their original counterparts at the same noise level. However, MANOVA. showed that for depth estimation of enamel lesions JPEG 53 resulted in a higher observer error. Conclusions: Both JPEG 53 and 27 could reduce some of the adverse effect of noise from the image by removing high spatial frequencies. JPEG 53, result ing in a compression ratio of 1:21, does not compromise the diagnostic perf ormance in general. JPEG 53 compression may however affect the ability to d etect enamel lesions.