PET/CT Image Fusion Headed for PACS
Financial Software to Integrate with RIS
Safe Harbors for IT
Reported at SIIM

PET/CT Image Fusion Headed for PACS

By Jerold W. Wallis, MD

With the growth of PET/CT comes an increasing demand for displaying fused PET and CT images on PACS monitors. Until now, this ability has been available on dedicated PET interpretation workstations, but it is rarely—if ever—present on general radiology PACS displays. Integrating the Healthcare Enterprise (IHE) has now created an Image Fusion Profile to help make this a more commonly available feature on viewing systems.

IHE works with vendors, users, and domain experts to create detailed guidelines on how to use existing standards—Digital Imaging and Communications in Medicine (DICOM), for example—so that different systems can interact smoothly to meet specific user needs. In this case, the result is the Image Fusion Profile. Vendors who choose to support the profile have the opportunity to do extensive testing with other vendors in a controlled environment, helping to ensure that the developed software will work well when it is released as a clinical product. Users can then ask for the profile by name when purchasing systems.

This profile addresses the ability to convey registered data from one system to another for further processing, storage, and display, as well as the ability to present repeatable fused displays consisting of a grayscale underlying image and a colored overlay image.

The profile has been designed for general radiology and other medical use, regardless of modality. At the same time, it is recognized that PET/CT will be the major use of the profile in the near-term, and it has been designed to meet the needs of PET/CT viewing on general radiology PACS viewing systems.

Systems that comply with the Image Fusion Profile have the ability to deal with a few special DICOM objects (see Figure 1). The profile would work for PET/CT data as follows:

  1. At your workstation, your software saves a Blending Presentation State, and possibly a Registration Object;
  2. When you send data to your PACS, both the image data and these two special DICOM objects would be sent for storage.
  3. When you view your images on the PACS display, you would select just the Blending Presentation State. The fused images would then automatically appear, with the data resampled as needed to account for the differences in image sizes and slice spacing.

Once the fused data appears on your viewing screen, you would have the ability to navigate through the fused volume, change window settings (using controls appropriate for PET or CT data), and adjust the transparency (weighting) of the two image sets. Based on our experience with PET/CT data, two additional requirements were included in the profile:

  • the ability to view tomographic images in multiple image planes (transaxial, sagittal, and coronal images); and
  • the ability to view the two image sets separately, side by side in a registered fashion, rather than just as a color-fused display.

Most likely, display systems that comply with the profile will allow the user to choose manually registered PET and CT images for display, even if the data come from an older system that does not supply a Blending Presentation State. Although the ability to perform quantitation on PET images to display Standardized Uptake Values (SUVs) would be of interest to some users, this ability is not required by the profile. Display of SUVs and other PET-specific features will likely find a place in a future IHE PET profile instead.

What is the next step? The Image Fusion Profile is now published for “trial implementation,” which means that vendors can review it and decide to implement test versions of their software. They will then be able to verify that their software works with other vendors’ data at a large multivendor testing session (connect-a-thon) to occur at RSNA headquarters in Chicago next January. Choosing to comply with and test the profile will allow vendors to be confident that their Image Fusion solution will work well and will meet users’ needs when it reaches the marketplace.

Vendors interested in the profile should contact the IHE about the participant workshops scheduled for this summer. A much more detailed document describing the profile is available on the IHE Web site (www.ihe.net). Portions of the profile are applicable to acquisition systems, processing workstations, image storage devices, and image displays, so inquiries from all are welcome.

In the meantime, users should start asking vendors, “Do you offer systems that comply with the Image Fusion Profile?” The fastest way to encourage vendors to supply the features of the profile is for many people to ask for it.

Jerold W. Wallis, MD, is an associate professor of radiology at the Mallinckrodt Institute of Radiology, Washington University School of Medicine, St Louis. He is chair of the Society of Nuclear Medicine DICOM Working Group and cochair of the IHE Nuclear Medicine Technical Committee.

Financial Software to Integrate with RIS

AMICAS Inc, Boston, has introduced the Vision Series Financials 7.0 financial software platform. This new version is designed to provide greater interoperability with the Vision Series RIS and Document Management programs offered by the company, as well as integration with AMICAS EDI Services. Vision Series features include electronic data acquisition, electronic claim submission, electronic claims remittance, system-generated explanations of benefits, online patient archival, automated collections module, comprehensive reporting, ad hoc reports and charts, integrated contract management, electronic statement processing, automated insurance follow-up, multi-corporation support, and tickler and internal mail capabilities. Version 7.0 results from the collaboration and direct feedback from the company’s financials customer base, and the AMICAS Customer Advisory Committee, which meets quarterly to help prioritize key feature development and guide the process. Vision Series Financials 7.0 is scheduled to be available in the second half of 2006. For more information, call (800) 490-8465 or visit www.amicas.com.

Safe Harbors for IT

Hospitals and radiology practices that have hesitated to equip referring physicians for PACS and electronic medical record (EMR) remote review for fear of running afoul of Stark anti-referral laws and Office of Inspector General (OIG) anti-kickback regulations might soon be given greater latitude if HR 4157 can make its way onto the President’s desk.

The Health Information Technology Promotion Act of 2005, recently passed by House Ways and Means Subcommittee on Health and sponsored by chair Nancy Johnson (R-Conn) and Rep Nathan Deal (R-Ga), specifies safe harbors for the provision of health information technology (IT) and training services to health care professionals.

The bill would amend Section 1128A(b) of the Social Security Act to allow an “entity” to provide health IT or training to a physician if made without condition that:

  • limits or restricts use of the IT only to patients receiving services at the entity;
  • limits or restricts use of health IT in conjunction with other health IT;
  • takes into account the volume or value of referrals or other business generated by the physicvian to the entity.

Former Health and Human Services (HHS) information technology chief, David Brailer, MD, PhD, on leaving office, alluded to the imminent creation of safe harbors aimed at accelerating the penetration of IT into the community to promote adoption of the EMR. “I would not be in my exit process if that was not over the hump,” he said in an April 25 press conference.

The annual meeting of the
Society for Imaging Informatics in Medicine,
April 27–30, Austin, Tex

Reported at SIIM

SCAR/SIIM to Certify Informatics Professionals.
The former Society for Computer Applications in Radiology (SCAR) has shed its old name for a new one, Society for Imaging Informatics in Medicine (SIIM). Also, the society announced that it is developing a certification program for PACS administrators, called Certified Imaging Informatics Professionals (CIIPs). SIIM is developing a test as well as criteria for test eligibility and has called for comments on its Web site (www.scarnet.net/ciip/). SIIM hopes to offer the first examination in September 2007.

Maximizing Workflow in DR. With 40% of hospitals reporting inadequate radiologic technologist staffing, and with general radiography average vacancy rates at two full-time employees, hospitals can ill afford inefficient processes in general radiography, according to session speaker Bruce Reiner, MD, director of research for the VA Maryland Healthcare System, Baltimore, and an associate professor with the Diagnostic Radiology Department at the University of Maryland School of Medicine.

Because technologists spend 33% to 57% of their time on quality assurance (QA), Reiner said that centralizing QA with “QA specialists”—whose job would include all QA tasks associated with DR—could greatly improve productivity. Other suggestions for achieving efficiencies included effective queuing that would redistribute inpatients to the slowest time of the day; new tracking technologies (such as RFID) and ubiquitous computing; a digital dashboard that can help identify bottlenecks and fallow periods; and continued automation, particularly in QA. “We need to reinvent QA in order to maximize workflow,” Reiner said.

Navigating the Tsunami. Eliot Siegel, MD, vice chairman and professor of diagnostic radiology, University of Maryland, and chief of radiology, VA Maryland Health Care System, led a session called Productivity in the Third and Fourth Dimensions: Getting the Most Out of Your 3D and Multiplanar Systems. “The widespread adoption of MDCT [multidetector computed tomography] has resulted in a crisis,” Siegel told attendees. “It’s a crisis in acquisition, image processing, and display.” Explaining that the average trauma scan at the University of Maryland Medical Center is 1,000 slices using a 1-mm reconstruction, he said, “It can take 5 to 10 minutes to just get the axial images from the raw data set. Now, we are being limited by the time it takes to reconstruct the images and send them.”

Three models are being used to handle 3D reconstruction of the data:

  1. Reconstruction at the scanner, which has the simplest workflow and can generate a bill, but it takes too much technologist time and is executed without radiologist input, Siegel noted.
  2. Radiologists perform the reconstructions, which slows down radiologist workflow.
  3. Implement a 3D laboratory, and a subspecialist technologist performs the reconstructions in the lab. It takes less tech and radiologist time, and it can generate a considerable amount of additional income, Siegel explained.

Studies that lend themselves to 3D reconstruction include CT angiography, neurovascular, aorta stent grafts, peripheral vascular, cardiovascular, tumor volume, CT urography, and virtual colonography studies. Siegel said that given current reimbursement, the best model is a combination of numbers 2 and 3. Be aware, though, he advised, of the following pitfalls:

  1. There is a tremendous amount of variability in quantitative measurements from one vendor’s workstation to another’s, so radiologists need what Siegel called “a reality check.” He noted, “There is a 40% to 80% variation on the estimation of carotid stenosis. Don’t put too much credence in automated measurements.”
  2. Bone removal software could remove a portion of the vessel, so be careful to balance scientific accuracy with artistic display.
  3. The old code CPT 76355 is no longer active, and, as of January 2006, 2-D reconstructions are considered a standard part of a CT procedure.
  4. The new codes in place—CPT 76376 (reconstruction at CT console; $10 professional component [PC]/$110 technical component [TC]) and CPT 76377 (postprocessing on an independent workstation supervised by a physician; $40 PC/$130 TC)—require medical necessity, and CMS has not yet published accepted medical criteria for medical necessity.
  5. The 3-D analysis should be reported in a separate session, clearly stating the reason for 3-D being added to 2-D. If an image is transferred to an independent workstation, be sure to obtain an order from the referring physician.

—C. Proval