Advanced Visualization in the EMR
Screen shot courtesy of TeraMedica.
The push to develop EMRs across healthcare enterprises really lies at the heart of  the Health Information Technology for Economic and Clinical Health Act (HITECH Act, part of the American Recovery and Reinvestment Act of 2009), and it is certain that all medical documents—including images and imaging reports—will eventually become part of the electronic health record. But those images are not just limited to diagnostic radiology. They include everything from endoscopy and cardiology images to  dermatology and dentistry. And they also include advanced visualization (AV) images.

What is the value of advanced visualization integration with EMRs? According to Janice Honeyman-Buck, PhD, associate professor and director of informatics, department of radiology, University of Florida, advanced visualization “has become a standard of care.” And while radiologists seem to own AV, she says, other specialists, particularly neurosurgeons and orthopedic surgeons, are increasingly relying on the technology.

“They want to be able to see the same thing the radiologists are seeing,” says Honeyman-Buck, who is also editor-in-chief of the Journal of Digital Imaging of the Society for Imaging Informatics in Medicine (SIIM), as well as an imaging informatics consultant. “So they needed to be provided with the tools they need to do their jobs.”

Gary Wendt, MD, professor of neuroradiology at the University of Wisconsin, Madison, says there are obvious benefits to being able to integrate advanced visualization into an EMR—but only after it has been extended throughout the enterprise. “Getting integrated advanced 3D imaging and perfusion processing into the EMR isn’t really worth the effort if your radiologists, neurologists and neurosurgeons can’t even get at the application at a regular workstation. If that’s the case, then why worry about it in an EMR?”

Consequently, before a healthcare enterprise even thinks about integrating advanced visualization into an EMR, it needs to see whether AV extends throughout its PACS, Wendt says. “And I don’t think most sites even have advanced visualization available everywhere within their PACS. We [the University of Wisconsin] have PACS integrated with EMRs at multiple sites and that’s really critical.”

One problem is the sheer difficulty of making sure that specific advanced visualization applications are available throughout an enterprise, says William Boonn, MD, chief of 3D and advanced Imaging and associate clinical director of imaging informatics, Hospital of the University of Pennsylvania in Philadelphia.

“The move to thin client helps considerably,” says Boonn, “But you still run into a problem that a particular application still needs to be installed everywhere you want to display the images. That’s typically not possible, or even reasonable, in a larger enterprise.”

According to Wendt, much depends on an enterprise’s PACS environment to successfully integrate AV into an EMR. Wisconsin uses McKesson PACS and Vital Images advanced visualization software, “so for us,” says Wendt, “once we had the integration of McKesson and Vital Images done, then it was pretty much of a no-brainer [to integrate the AV into the EMR] because if it’s available on the PACS then it’s available within the EMR.”

With other systems, it could be more problematic, says Wendt, because the process of integrating advanced 3D capabilities on a radiologist’s workstation “doesn’t necessarily mean that the advanced visualization will be available on the web interface or within the interface of the EMR.”

According to Wendt, much depends on the PACS architecture and how vendors handle their advanced integration. “With us—with the McKesson/Vital Images integration—it really doesn’t make a lot of difference whether you’re on a stand-alone [workstation], if you’re on a thin-client radiology workstation, on a web-based PACS workstation, or if you’re in the EMR integration, because it all behaves the same.”

Who will use it?

Joseph Marion, founder and principal of Healthcare Integration Strategies, a consulting firm located in Waukesha, Wis., says there are three levels of advanced visualization user. The most advanced is the diagnostician who interprets the information to create the report, while the most “casual” will likely be the referring physician who is strictly looking for results and whose need to access images may be fairly minimal. Then there’s the in-between user—the specialist such as the neurosurgeon or orthopedic surgeon—whose requirements are somewhere between that of the diagnostician and the simple viewer.

It is this latter group that Wendt expects will make the best use of advanced viz tools in an EMR. “Realistically, doing things like 3D and 4D image processing, perfusion and cardiac functional, that really isn’t going be aimed at your primary care physician,” Wendt says. “I don’t think they are going to be logging on to an advanced 4D perfusion scan and doing perfusion processing on the brain.”

Instead, these end-users will use the EMR to access results quickly. “You get a primary [care] physician taking a quick look at a chest x-ray for example. These docs aren’t going to be playing in the EMR environment, at least not quite so heavily.” But, Wendt can see specialists such as interventional neuroradiologists, stroke neurologists and neurosurgeons interacting with images via the EMR. “Someone like an orthopedic surgeon who wants a curved MPR,” he says.

While the integration of medical images, including advanced visualization images, into EMRs is addressed in the meaningful use criteria within the American Recovery and Reinvestment Act (ARRA), the requirements that will specifically impact medical imaging—medical device interoperability and multimedia support—don’t go into effect until 2015.

That’s not soon enough, according to Joseph Marion, founder and principal of Healthcare Integration Strategies, a consulting firm located in Waukesha, Wis., and Janice Honeyman-Buck, PhD, associate professor and director of informatics, department of radiology at the University of Florida. “It’s like building a house by attaching the siding before putting up the frame,” says Marion. “There are a lot of things that have to get done to get to an electronic health record, but imaging, is a key part of that and can’t be ignored.”

Honeyman-Buck agrees, adding that despite the fact that meaningful use requirements seem to have put medical imaging at the tail end of the process, healthcare administrators need to begin addressing imaging now.

If not, says Marion, then “three years from now when people get around to addressing images, if they are making their capital plans based on those ARRA priorities, then they’ll end up with a solution that doesn’t support images. And then they may be back square one.”

Instead, healthcare enterprises need to decide now what their strategy will be for including images into the EMR, says Marion, adding that if a hospital has a number of different vendor systems deployed requiring different image viewers, then it’s not going to work, “because there’s no doctor out there who’s going to want to learn how to use a bunch of different viewers.”

Marion is a proponent of enterprise image management, where an enterprise will have its imaging PACS feed all information into a centralized archive. “That becomes the gatekeeper for the EMR,” says Marion, “and then I have a consistent way of addressing that information to the EMR.”

Barriers of entry

At the University of Pennsylvania, EMR interoperability and image integration is a “work in progress,” says Boonn. “We actually do have a certain level of EMR integration. Users of the EMR are able to view images and results—it’s something we’ve been able to do for years. But the ability to do dynamic manipulation of 3D data sets, that part is more troublesome.”

An advanced visualization product such as TeraRecon’s IN- tuition that through the web allows the clinical user to view and dynamically manipulate images, “is a tremendous advance,” says Boonn. “So we have the technology to manipulate these data sets, it’s just a question of getting everything implemented.”

Healthcare enterprises looking to integrate advanced visualization into their EMR need to examine the kind of environment their EMRS are running in, Wendt recommends. “If you are using thin-client environments like Winterm or Citrix emulation environments with really low-cost or low-end PCs, will you have good enough displays to even handle advanced visualization?”

“With the current advanced 3D processing, users don’t need as much CPU and GPU capabilities because they can be offloaded to the back end of the advanced visualization product, says Wendt, but a ‘reasonable’ display is still needed.”

“If you are running on a Winterm or Citrix box that gives you 800 x 600 [monitor] resolution, and you are locked to that, you are not going to have real good experience,” says Wendt. “Even if you offload all the processing, you still need to be able to present those images in a reasonable resolution and bit depth in order to use the tools.” And while with the new advanced visualization products you no longer need a lot of RAM, GPU and network bandwidth to deploy the AV, he adds, “you still need an adequate desktop or window into the world.”
Finally, says Wendt, if a facility or health system is going to change over to an enterprise 3D model, it needs to be cognizant of what its end-users and advanced-users want. “If 95 percent of your people are end-users, and 5 percent are advanced, make sure you don’t drown those 95 percent who are end-users in [advanced technology] they have no interest in.”

Avatar—not the movie

Researchers in Europe have taken somewhat of a different approach to advanced visualization and EMRs.
Using an avatar—a 3D representation of the human body—researchers from IBM and Nhumi Technologies in Zurich have created a software visualization tool that allows doctors to “visualize” patient records. Andre Elisseeff, of Nhumi Technologies, one of the inventors of the software tool, compares it to “Google Earth for the body.”
Using this tool, clinicians can simply click on a particular part of the 3D image to trigger a search for the medical records associated with that part of the body. For example, if a patient comes into a doctor’s office with a complaint of back pain, the doctor can simply click on the area representing the spine and see all of the information relating to the spine, including lab results, text entries and medical images.

A screenshot of the visual patient record displays basic patient information such as a social security number and name and address on the right panel, while the 3D anatomical model—displayed on the left panel—contains arrows indicating areas of the body that have been treated. By clicking a mouse on an arrow or body part on the left panel, a clinician can retrieve the pertinent medical information on the right panel, which also contains the patient’s medical record. In addition, the clinicians can manipulate the 3D image—or a selected part of that image—to get different anatomical views of the cardiovascular system, the central nervous system, or the muscular system.

“In the hope of speeding the move towards electronic health records,” says Elisseeff, “we’ve tried to make information easily accessible for healthcare providers by combining medical data with visual representations, making it as simple as possible to interact with data that can improve patient care.”

The tool was tested last year at the Thy-Mors Hospital, a facility in northern Denmark with about 11,000 patient beds and more than 65,000 outpatient visits per year. It allowed Hardy Christoffersen, MD, head of the hospital’s surgical outpatient clinic, to “see more information than just what the patient tells me is bothering him or her that day—information for which I would otherwise have to spend considerable time searching through our current records system. “ According to Nhumi Technologies, clinicians at the hospital testing the system were able to reduce by two-thirds the time it usually takes them to access patient medical histories.

Christoffersen also found that the visualization system provides other benefits as well. For example, it helped doctors spot information that was indirectly related to current health problems but was still relevant to treatment.
And as is typically the case with 3D imaging, the tool also, Christoffersen says, facilitates doctor-patient communication by giving the patient an understandable way of visualizing his or her body and health issues.