How will DICOM accommodate MR advances?

As data from MR acquisitions reach the gigabit level, image management techniques must meet the challenge. An evolving DICOM standard provides one answer.

At the plenary presentation of the recent medical imaging meeting in San Diego, Dr. Chuck Mistretta, a professor at the University of Wisconsin and a well-known expert in this field, described some of the impending advances in MR. We are only at the beginning of a new era of innovative acquisition and reconstruction techniques in k-space, the MR frequency data domain, according to Mistretta.

What frightens implementers is that the data researchers are talking about run into gigabytes for a single acquisition and there could be several acquisitions for each patient. Clinical benefit will no doubt be considerable, especially for the angiography acquisitions, but how are we going to accommodate these humongous data sets?

Life would be boring without challenges, and the DICOM standardization committee has taken up this one, with an effort to improve the "MR object," its definition for the way MR data are handled. Currently, individual MR images are sent as separate entities. The receiver software has to collect them, look at each individual "header," and figure out how they relate.

Another construct in DICOM, the "multiframe object," groups a complete set of images and allows them to be moved in a single transaction. This object is used by some of the cine-like modalities, such as ultrasound, angiography, cardiology, and nuclear medicine. An advantage to multiframe object is that a receiver has to look only in a single header to learn the difference between the individual frames, such as acquisition modes. That, in turn, allows a quick sorting to accommodate a particular hanging protocol that is needed for a specific application and/or physician.

The new MR object accommodates the multiframe construct and can exchange information that until now has been sent only in a proprietary format, such as spectroscopy data. It will also allow the exchange of "raw data," so that different image processing techniques can be used at a specialty workstation or by researchers. The proposal is out for public comment; interested parties can evaluate it and send in their comments (www.nema.org/dicom).

The new MR object addition to the DICOM standard will probably be finalized within the next six months.

What does this mean for MR users and buyers? First, it is always a good idea to specify in an RFP or purchasing contract the requirement to upgrade the equipment for a reasonable fee (or, even better, no fee). This does not mean that all equipment will need to be upgraded, but you might consider this feature if you want to have access to information such as spectroscopy in a nonproprietary format and to be ready for efficient exchange of data gathered with new acquisition methods.

People often ask me when DICOM will finally be completed and no longer evolve. The truth is that as long as modalities evolve, and researchers come up with new techniques, changes in the interface will always be needed.

It's best to be prepared to deal with these changes. MR is not the only modality that is changing. The good news is that most new techniques improve patient care by offering procedures that are less invasive and less risky than those available.

Mr. Oosterwijk has participated in the DICOM standardization process since its inception as member of several working groups. Send questions or concerns to herman@otechimg.com or visit his Web site at www.otechimg.com for more standards news.