SIIM University at SIIM 2007 in Providence, R.I., provided updates on digital mammography on Saturday. Margarita Zuley, MD, University of Rochester, said it has been an active year for the modality.
For starters, acquisition units have experienced a lot of change, such as advances in detector technology, targets and filters, matrix imaging processing and patient throughput.
One major challenge in digital mammography is the fact that radiologists use the patient herself as a control and look for changes over time. That means that technological changes over a few years really make it difficult to determine whether something suspicious is a meaningful change in the woman. Zuley said that “in the future, we might use different algorithms for different tissue types or different lesions.” Right now, the processing algorithms is the “biggest conundrum we’re dealing with.”
Zuley described two classes of problems facing digital mammography:
- 1. Content display. Differences among vendors made it difficult for mammography readers to make accurate assessments. However, the Mammography Image Profile was finalized in 2006 and has been very well adopted, she said. The profile prescribes that everything be scaled off pixel pitch as well as guides how vendors display pixels. It also prescribes that the scaling is accurate for any magnification.
- 2. Workflow constraints. IHE just starting to work on this, Zuley said. Digital mammography involves different user interfaces from analog mammography. How a facility sets up their workflow determines maximum efficiency. The modality offers display options and the need to develop task-oriented hanging protocols. She said that the literature is now starting to cover screening efficiencies.
Richard Morin, PhD, discussed new practice guidelines for digital mammography from the American College of Radiology. He said the guidelines are not enforceable rules of practice but more of an educational tool to assist with providing appropriate radiological care.
Morin said that a guide was needed because of the industry lack of understanding and knowledge. Plus, government agencies and manufacturers had no consensus document. There was a need to provide technical details for basic guidance. The guidelines were the result of several societies: the ACR, SIIM, RSNA and AAPM. The greater the range of people involved, the broader the range of expertise included.
The guidelines have a multifaceted focus, including digital mammography image acquisition, display and archiving/storage. They were originally published in the Journal of the ACR in August, 2006 with vendor feedback accepted from August to December, 2006. After field review, final guidelines were ready for ACR approval last month and ACR membership approved them last month as well.
Morin said that image quality needed to be covered because it is so different from other modalities. Another reason guidelines were needed is because digital mammography is being adopted so rapidly. “Changes will occur and the need to adopt content frequently is recognized,” he said. He added that any and all feedback is welcome and appreciated.