Medical Image Management: Bridging the DICOM Divide
Evolving diagnostic technologies combined with expanding health provider markets has created an abundance of digital information. The digital imaging and communications in medicine (DICOM) format helps standardize this resource to deliver information interoperability. The challenge? To ensure accuracy and consistency of medical care, DICOM files are much larger than other image formats.
Compression is therefore critical to deliver widespread interoperability, but consistency is key to ensure no essential data is lost in translation when files are compressed, decompressed, and viewed. Add in the lack of native support for this file type in popular operating systems, and the pain points of this imaging process are problematic for its prognostic potential. Here are three ways organizations can bridge the DICOM divide and deliver improved digital outcomes.
The Accessibility Issue
As noted by NCBI-hosted journal research, it’s not always easy for staff to access DICOM files on-demand because “in contrast to other image formats such as JPEG or TIFF, the individual DICOM files are not recognized by Windows as image files.” This is further complicated by the use of physical media — such as compact discs (CDs) — to transport large DICOM files.
Since operating systems offer no native support for DICOM files, CDs must include viewer software; if viewers are hidden in subdirectories or experience compatibility issues with specific desktop configurations, staff spend more time searching for image solutions than solving medical issues. Here, integrated viewer APIs can help reduce accessibility barriers by equipping current medical applications with the ability to identify and open DICOM files on-demand.
Compressed to Impress
DICOM files differ from other image formats because key information is grouped into datasets including patient ID, name, and image pixel data that comprise the larger DICOM object. Each object can only have one attribute containing image data, but these attributes can contain multiple “frames” to create multi-dimensional images of specific scan types. In addition, DICOM uses three Data Element encoding schemes depending on data type.
As a result, the size of DICOM files can quickly exceed digital transfer limits for common communication channels such as email. For example, PET scans are often 30MB or larger, while mammography data can surpass 45 or even 50MB. In this case, compression is key; if file sizes can be reduced, it’s possible to limit the use of physical media, which remains a top complaint among picture archiving and communication systems (PACS) users.
To ensure effective compression (and decompression), however, organizations must leverage tools that are non-destructive. DICOM data sets are naturally interrelated. If any part of the is file destroyed and reconstituted during the compression/decompression process, the format becomes unusable.lines. Here, a single-application approach to invoice processing helps improve invoice routes and reduce redundant approval steps.
According to a recent DICOM strategic document, “the pressing needs in DICOM (as indicated by the priorities of the various working groups) are to address issues relating to new modality technology, structured and coded documents for specific clinical domains, workflow management, security, and performance.”
Workflow management is a particularly pressing concern, since DICOM formatting supports multiple image types including lossy/lossless sequential JPEGs, JPEG 2000, JPEG-LS, and JPEG-XR for common images such as MRIs or CT scans. In the case of JPEG 2000, for example, organizations need a way to convert 48-bit color space into standard 24-bit without losing image quality. This requires decoding functions capable of recognizing the separate memory buffers for RBG data in planar image format and then reducing these samples into 8-bit values interleaved in BGR order to allow storage as a standard Windows bitmap. To bridge the DICOM divide, organizations need tools that empower easy access, deliver non-destructive compression, and are capable of decoding multiple DICOM image types.