Related Experiment Videos
Application of wavelet compression to digitized radiographs
M A Goldberg1, M Pivovarov, W W Mayo-Smith
1Department of Radiology, Harvard Medical School, Boston, MA.
AJR. American Journal of Roentgenology
|August 1, 1994
Summary
Wavelet-based compression is effective for digitized radiographs, maintaining diagnostic quality up to a 30:1 ratio. This technology shows promise for teleradiology and Picture Archiving and Communication Systems (PACS).
Area of Science:
- Radiology
- Image Processing
- Digital Imaging
Background:
- Image data compression is crucial for teleradiology and Picture Archiving and Communication Systems (PACS).
- Wavelet compression is a novel technique with limited reported studies in radiology.
Purpose of the Study:
- To demonstrate the application of wavelet-based compression technology to digitized radiographs.
- To evaluate the diagnostic significance of image degradation at various compression ratios.
Main Methods:
- Twelve abnormal radiographs were digitized, compressed (10:1 to 60:1), and decompressed using a wavelet-based lossy algorithm.
- Seven radiologists assessed paired original and compressed images for clinically relevant degradation on a high-resolution workstation.
- Quantitative error measures were recorded alongside subjective assessments.
Main Results:
- No clinically relevant image degradation was found below a 30:1 compression ratio.
- Skeletal radiographs showed a trend towards greater sensitivity to compression, though not statistically significant.
- Quantitative error measures increased progressively with higher compression ratios.
Conclusions:
- Wavelet-based compression techniques show promise for digitized radiographs due to image quality and computational efficiency.
- Further observer performance studies are warranted based on these initial findings.