The effects of technical factors on the fractal dimension in different dental radiographic images

PURPOSE: The aim of this study was to assess the impact of exposure parameters and image formats on fractal dimension (FD) values in periapical, panoramic, and CBCT images. MATERIALS AND METHODS: Seven dry male mandibles were selected, and a Gutta-Percha was used to identify identical regions of interest. A periapical radiograph was taken with 60 kVp/7 mA and exported in DICOM, JPEG, TIFF, and PNG formats. Nine periapical radiographs (60, 65, 70 kVp; 4, 5, 6 mA) were taken from seven dry human mandibles. Additionally, 12 panoramic radiographs (60, 70, 81, 90 kVp; 5, 8, 13 mA) and 10 CBCT images (with different scanning options and FOVs) were taken from each mandible. FDs were measured from a standard area. RESULTS: The intra-class correlation coefficient demonstrated a high degree of agreement between observers. No significant difference was found between TIFF and PNG formats (p > 0.05). The highest FD mean was found in TIFF format, while the lowest FD mean was found in JPEG format (p<0.001). There was no significant difference between kVp and mA settings in periapical images. In panoramic images, a significant difference was found at 90 kVp (p = 0.001) and 13 mA (p<0.001), with lower FD values observed at these settings. There was no significant difference between FOV and resolution in CBCT images (p > 0.05). CONCLUSION: The format of the image can influence FD. For periapical and panoramic radiographs, kVp and mA settings do not have a significant impact on FD. However, fractal analysis may not be an ideal method for evaluating three-dimensional images, such as those obtained with CBCT.