Swelling archives warrant closer look at compression

Image compression is broadly categorized as lossless or lossy. With lossless compression, a compressed image can be decompressed and displayed as an exact digital replica of the original. With lossy compression, redundant pixel data are discarded during the compression process so that the compressed image is only an approximation of the original, therefore it cannot be returned to an original state. Although both types of compression are commonly used within the medical imaging community, institutions are much more likely to depend upon lossless compression for diagnostic purposes, even though lossy images, saved as a fraction of the original file size, are often diagnostically equivalent. With more and more digital modalities coming online and image studies growing ever larger, institutions failing to take full advantage of lossy compression are missing an opportunity to slow the growth of their image archives and IT infrastructure costs. Today, most PACS vendors include some form of image compression technology within their product offerings. Of 13 vendors sampled during an informal survey in June 2003, eight employ JPEG 2000 compression, which was incorporated into the DICOM standard in 2001, while the remainder use wavelet compression, which is the underlying methodology used in JPEG 2000. JPEG 2000 is an industry standard that enables image sharing across platforms and product lines. It also provides a single mechanism for creating lossless and lossy images, and gives institutions the flexibility to apply unique rates of compression to individual images based on modality, patient history, image size or other factors. Although it can be used judificiously to great advantage, compression is most often applied in an "across the board" manner to all images. The incorporation of JPEG 2000 within the DICOM standard does little to guarantee its longevity, or the quality of every JPEG 2000 implementation. At some point in time, especially as the demand for more compact, higher-quality lossy images grows, institutions will begin experimenting with and employing more advanced compression methods.