Biostatistics Research Today is a free monthly online journal that collates and summarizes the latest research about Biostatistics, including details on statistics, uncertainty, probability, modeling.

Direct digital lateral cephalometry: the effects of JPEG compression on image quality.

Wenger NA, Tewson DH, McDonald F

Department of Oral Health, West Block, Norfolk and Norwich University Hospital, Colney Lane, Norwich, UK. nicholas_a.wenger@nnuh.nhs.uk

OBJECTIVES: This study used an aluminium test object to assess the effect of the Joint Photographics Expert Group (JPEG) compression algorithm, on direct digital cephalometric image quality. METHODS: The aluminium block of 15 steps, with 20 holes in each step was radiographed in a Planmeca Proline 2002 digital cephalometric machine with Dimaxis2 software. Six different JPEG compression ratios were used to capture the cephalometric images. These ratios were 60%, 70%, 80%, 90%, TOP QUALITY JPEG (TQJPEG 98%) and TIFF (uncompressed). The images were taken at 68 kV and 12 mA with a 7 s exposure. Six experienced observers viewed the monitor displayed images, which were presented randomly. This was repeated one month later. The number of holes detected by each observer was plotted against each compression ratio. Intra-observer and inter-observer reproducibility was calculated using the Mann-Whitney U-test. Differences between the compression ratios were assessed using a Kruskal-Wallis one-way analysis of variance. RESULTS: When comparing intra-observer reproducibility, it was found that there were only four of 36 comparisons that showed a statistically significant difference (Observer 1: 60% (P=0.004), TQJPEG (P=0.019); Observer 2: TIFF (P=0.005); Observer 3: 90% (P=0.007)). Statistically, there was no significant difference with inter-observer reproducibility. There was no statistically significant difference between the image quality obtained from each compression ratio. CONCLUSIONS: The results showed that JPEG compression does not have any effect on the perceptibility of landmarks in the aluminium test object used in this study.

Published 15 May 2006 in Med Eng Phys, 28(6): 560-7.
Full-text of this article is available online (may require subscription).

© 2005-2008 Biostatistics Research Today. All Rights Reserved.