Strengths, Struggles & Strategies Managing Image Data

Strengths, Struggles & Strategies Managing Image Data

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon

Onco-Quant-Gallery-PETCT.jpg

Source: Onco-Quant-Gallery-PETCT.jpg - Image Data Management
Leveraging image data for more precise treatment in oncology and beyond.

The drive for heterogeneity in molecular imaging calls for management of complex imaging data in massive volumes, a task that is challenging information technology experts and medical community members alike. As the demand for image data management deepens in areas like molecular imaging and genomics research, how is the latest technology influencing medical imaging informatics?

The drive for heterogeneity in molecular imaging calls for management of complex imaging data in massive volumes, a task that is challenging information technology experts and medical community members alike. As the demand for image data management deepens in areas like molecular imaging and genomics research, how is the latest technology influencing medical imaging informatics?

Assessing the advancements

Harnessing the huge amount of data that are required in molecular imaging, genomics research and clinical practice is a complex and daunting goal. Numerous technological advancements have recently been made, but much of this has come along with the digitization of technology and the advent of online and cloud platforms, creating even more gargantuan volumes of data requiring increasing numbers of processing servers.

Regionalization is another vital area of growth in image data management. “There is a more regional approach to data warehousing, as we move away from the world in which three different solutions were previously used by organizations to safeguard and store one piece of data,” comments David Price, MRT, director of diagnostic services at Queensway Carleton Hospital in Ottawa, Canada. Online storage is now very inexpensive, making the near line component of warehousing obsolete.

The advent of diagnostic imaging repositories also has changed the game, explains Price. “There are now five diagnostic imaging repositories in our entire province which act as long-term archiving for every hospital in the area,” he says. “In our region, all 40 hospitals have access to everyone’s images, which is especially helpful when patients are being transferred or moving. The practice pattern has changed because of alterations in storage. This initial storage problem has ultimately improved patient care.”

Michael Minear, chief information officer for the UC Davis Health System and Center for Molecular and Genomic Imaging in California, adds to the idea of repositories: “The trend is to have these images now, which many places are accomplishing by creating meta-repositories. How you manage and provide access to them, however, is the biggest issue.” Minear says that these repositories have the potential to aid in tracking and managing radiation dose amongst patients.

At Minear’s organization, “second generation” image data management has emerged over the last five years. Their three PACS have been integrated with the EHR, allowing for any relevant images from patients to have a section of thumbnails that is then auto-linked to other systems for storage.

Moreover, “the advent of vendor neutral archives is certainly something that people have been talking about in the past couple of years,” says Kenneth C. Wang, MD, PhD, of the Baltimore VA Medical Center in Maryland. “Vendor neutral archives change the historical ways that practices handle their data.” Data are no longer contained in a system owned by one department, and storage systems are no longer provided by single vendors.

Those who interact with imaging data also can utilize very thin clients or zero footprint viewers. “This is changing the way that users throughout clinical enterprise may access information and also alters who can view the data,” he says. However, this technology introduces issues of security, ownership and medical information governance.

“The shift from paper to electronics hasn’t changed our premise of safeguarding,” explains Price. “This change hasn’t altered the fact that if information is lost, a patient’s care is compromised. It has, however, made it easier for more problems to occur.”

Lastly, Wang refers to the increasing availability of software platforms for performing analytics throughout clinical enterprise. Tools to devise and access data that were once difficult to obtain and use have now become user-friendly and are more widely accepted.

Moving advancements into action

The host of technological advancements in image data management is having a collective impact on how data are being used. Vendor neutral archives, zero footprint viewers and analytic software platforms are contributing to increased democratization of imaging data, according to Wang. More democratization means increased opportunities for innovations in research and patient care.

Beyond democratization, the relationship between diagnostic imaging and information technology has evolved. “Diagnostic imaging used to own all of data management,” explains Price. “In reality, it doesn’t need to own any of it. It makes more economic sense to turn IT into a service provider for imaging. We need to manage our care and we need to know that we have a very tight relationship with internal and external IT providers to ensure that they continue to meet our needs.” The capability of experts to service across organizations has rapidly progressed, making the IT world simpler and cheaper as those in imaging relinquish control and leave it to the experts to handle.

As these technological advancements move into application, challenges that accompany said instances of progress should not go unnoticed. First and foremost is the size of the raw data and the need to structure it. Questions abound as to the best ways to capture, store and move data. “The tools we’re currently using aren’t scaled to handle that research,” says Minear. “We have technological capabilities like sequencing, but the other components are still lagging behind.”

In spaces where genomic data are used, many are left wondering how to link that information with the EHR. At UC Davis Health System and Center for Molecular and Genomic Imaging, mice are inserted with portions of tumors from patients with bladder cancer in order to determine the effectiveness of treatment. Patient cases and those of the mice must be tracked, which produces an enormous amount of data. As the push for personalized medicine continues in the medical community, considerations must be made in terms of how to make these data applicable and accessible to patients.

“To find what’s relevant for a specific patient case, part of the challenge that comes with visualization of data management is adding cohort dimensions,” remarks Minear. “We have built nine registries on one platform so that we are not only building populations, but integrating and linking them as well. We need to know these connections to really understand the clinical problem.”

The road ahead

Though much work is still needed to fully optimize the methodology used to manage a virtual storm of data that is both accessible and scalable, strategies are in place to reach this goal.

Finally, a key way to make information exchange and image data management more practical is to spread awareness about semantic interoperability.

“This is important for users to be aware of when making decisions about which systems to purchase for a department or working with IT staff,” says Wang. “Operational decision making today has an important impact on our ability to achieve semantic interoperability in the future.”