| Abstrakt: |
Objectives: To evaluate the influence of DICOM and JPEG formats on the reproducibility of identification of cephalometric landmarks on posteroanterior digital radiographs. Methods: The sample consisted of digital images of posteroanterior radiographs saved in DICOM format; they were subsequently converted into JPEG format with quality factors of 100, 80, and 60. After sample blinding, 18 cephalometric points were identified on each image by three calibrated orthodontists using a system of x- and y-axis Cartesian coordinates. The intraclass correlation test was applied to analyze the occurrence of systematic errors, and Dahlberg's formula was used to calculate casual errors. The reproducibility between the DICOM and JPEG formats was compared by analysis of variance. All tests were conducted with a significance level of 5 %. Results: Each examiner repeated the procedure 30 days after the initial test with 20 % of the images, which were randomly selected, in each file format. We found that the cephalometric landmarks in the posteroanterior digital radiographs presented intraexaminer and interexaminer reproducibility; the exceptions were points ZL, ZR, AZ, JR, NC, and CN in the y-axis and A6 in the x-axis, which were due to intraclass correlations under 0.90 and casual error over 1.0 mm, irrespective of file format (DICOM or JPEG). Conclusions: The DICOM and JPEG file formats and quality factors of 100, 80, and 60 did not affect the intraexaminer and interexaminer reproducibility of cephalometric landmarks. [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink