RSNA 2004 

Abstract Archives of the RSNA, 2004


SSM21-02

Thinner but Less Compressible: Effect of Slice Thickness on Image Compressibility

Scientific Papers

Presented on December 1, 2004
Presented as part of SSM21: Radiology Informatics (Internet 2, Image Compression)

Participants

Khan Mohammad Siddiqui MD, Presenter: Nothing to Disclose
Eliot Lawrence Siegel MD, Abstract Co-Author: Nothing to Disclose
Bruce Ian Reiner MD, Abstract Co-Author: Nothing to Disclose
Amy Elizabeth Musk MD, Abstract Co-Author: Nothing to Disclose
Jeffrey Johnson PhD, Abstract Co-Author: Nothing to Disclose
Olivier Crave PhD, Abstract Co-Author: Nothing to Disclose
Mariappan S. Nadar PhD, Abstract Co-Author: Nothing to Disclose
et al, Abstract Co-Author: Nothing to Disclose

PURPOSE

With the widespread adoption of multi-slice CT (MSCT) scanners, the number of images generated has increased substantially due to a significant decrease in the average slice thickness. Effect of thinner slices on image compressibility has not been assessed. The purpose of this study was to assess the effect of slice thickness on JPEG2000 compression

METHOD AND MATERIALS

Five thoracic CT data sets obtained using a MSCT scanner at a collimation of 0.75mm (kVp 120, mAs 90) were reconstructed at various slice thicknesses (0.75mm, 1.5mm, 3mm, 6mm and 10.0mm). JPEG2000 compression (4:1, 6:1, 8:1, 12:1, 16:1, 32:1, and 64:1) were applied to these datasets. The effect of compression at mediastinal (350/50) and lung (1500/-600) window and level settings was also assessed. Image quality was measured utilizing Peak Signal to Noise Ratio (PSNR) and the Sarnoff JNDmetrixTM Visual Discrimination Model

RESULTS

Using the standard 2D JPEG2000 compression, the thinnest (0.75 mm) slices were substantially less compressible than thicker (up to 10 mm) slices for the same level of image quality (i.e., constant JND or PSNR value), particularly at higher compression ratios. For images evaluated with the mediastinal window settings and compressed to a level where the JND was 2.0, the ten mm slices could be compressed at a ratio of 8:1 in comparison to only 6:1 for 0.75 mm slices. At a JND of 4, the 10 mm slices compressed at a ratio of 16:1 whereas the 0.75 mm slices were only compressed at a ratio of 12:1. Increasing the 2D compression ratios of the 0.75 mm slices from 6:1 to 8:1 and 12:1 to 16:1 raised their distortion levels significantly above those of the 10 mm slices by as much as 1 JND or 8 dB in PSNR

CONCLUSIONS

Current literature and accepted practice suggest that compression ratios using 2D JPEG2000 can be as high as 8:1 for CT without visually perceptible loss of image quality. However our research suggests that compression of MSCT thinner sections at this ratio of 2D JPEG 2000 would likely result in substantially greater and potentially clinically significant, perceptible degradation of image quality and that a lower compression ratio should be utilized for these thin slices

DISCLOSURE

J.J.,M.S.N.,O.C.: work for siemens corporate research

Cite This Abstract

Siddiqui, K, Siegel, E, Reiner, B, Musk, A, Johnson, J, Crave, O, Nadar, M, et al, , Thinner but Less Compressible: Effect of Slice Thickness on Image Compressibility.  Radiological Society of North America 2004 Scientific Assembly and Annual Meeting, November 28 - December 3, 2004 ,Chicago IL. http://archive.rsna.org/2004/4417613.html