OBJECTIVE: The purpose of this study was to compare dimensions on charge-coupled device-acquired images under various enhancements with dimensions on film radiographs and in vivo dimensions with respect to the linear measurement of periradicular radiolucent lesions. STUDY DESIGN: The dimensions of 25 lesions imaged by means of a charge-coupled device-based digital receptor and Ektaspeed Plus radiographic film were measured and compared with dimensions as actually measured on impressions of the lesions taken during surgery. Digital images were displayed in unenhanced, equalized, and equalized/color-coded modes. The color-coding protocol was repeated to determine the method error, and each image treatment was also remeasured to determine intrarater reliability. Differences were compared through use of a Friedman 2-way analysis of variance with a follow-up Wilcoxon signed rank test (alpha = 0.01). RESULTS: Actual lesion dimensions were larger than dimension estimates obtained with digital images (range, 23% to 35%) and film radiographs (range, 29% to 43%). Color-coded images were significantly less accurate than equalized images and unenhanced images. Images equalized through use of the measurement algorithm of the resident software provided estimates that were more accurate than estimates made with film radiographs and a standard millimeter rule. Intrarater variability was low. The application of color coding was found to be unreliable. CONCLUSIONS: When applied to intraoral images, color-coded image processing of digital images had limited value in the estimation of periradicular lesional dimensions.
OBJECTIVE: The purpose of this study was to compare dimensions on charge-coupled device-acquired images under various enhancements with dimensions on film radiographs and in vivo dimensions with respect to the linear measurement of periradicular radiolucent lesions. STUDY DESIGN: The dimensions of 25 lesions imaged by means of a charge-coupled device-based digital receptor and Ektaspeed Plus radiographic film were measured and compared with dimensions as actually measured on impressions of the lesions taken during surgery. Digital images were displayed in unenhanced, equalized, and equalized/color-coded modes. The color-coding protocol was repeated to determine the method error, and each image treatment was also remeasured to determine intrarater reliability. Differences were compared through use of a Friedman 2-way analysis of variance with a follow-up Wilcoxon signed rank test (alpha = 0.01). RESULTS: Actual lesion dimensions were larger than dimension estimates obtained with digital images (range, 23% to 35%) and film radiographs (range, 29% to 43%). Color-coded images were significantly less accurate than equalized images and unenhanced images. Images equalized through use of the measurement algorithm of the resident software provided estimates that were more accurate than estimates made with film radiographs and a standard millimeter rule. Intrarater variability was low. The application of color coding was found to be unreliable. CONCLUSIONS: When applied to intraoral images, color-coded image processing of digital images had limited value in the estimation of periradicular lesional dimensions.
Authors: Abhishek Ranjan Pati; Vijayalakshmi K R; Sanat Kumar Bhuyan; Rajat G Panigrahi; Smita R Priyadarshini; Satyaranjan Misra; Chandravir Singh Journal: J Clin Diagn Res Date: 2014-11-20