The concept of a shared central repository has been a staple of information technology for decades.
Diagnostic imaging is arguably the most important technology ever created for patient diagnosis and treatment. Utilization of more than 700 million tests per year in the United States seems to underscore its universal acceptance. Medical images are typically large files, accounting for more than 80 percent of the volume of a patient's electronic record. Imaging is used to support 60 percent of all patient diagnoses. For clinicians, it is critically important to view not only a current image, but also prior images in order to track the trajectory of the patient's condition. As imaging technology has evolved, so too has storage technology. Yet do most imaging professionals understand where images are stored and how storage impacts capabilities for image sharing?
Traditionally, diagnostic images have been created at a departmental level and stored in archives that are proprietary to that department. In most environments, radiology, cardiology and orthopedics each has its own image repository, and images are not shared across archives. The complexity is compounded with some departments having multiple picture archiving and communications systems (PACS) from various vendors.
Today, image silos are commonplace. Unfortunately, this model does not support a patient-centric view of images, and it does not support the patient as they traverse the healthcare ecosystem. A sad result of this fragmented imaging infrastructure is that 50 percent of certain scans are estimated to be duplicate tests. The results are often delayed treatment, added costs and unnecessary exposure to radiation. The healthcare industry has a compelling need for image interoperability between departments, sites and enterprises. Today, we are seeing early adopters address this problem with new technology.
Many innovators have pursued a strategy that allows for storage of all diagnostic images in a single, shared repository. The concept of a shared central repository has been a staple of information technology for decades. It has been a fundamental approach for the collection and sharing of objective patient data for a similar period of time. For today's approach with image storage, perhaps the time has come for a trip back to the future of information technology strategy.
Within healthcare, the centralized imaging repository is often referred to as a vendor-neutral archive (VNA). Consolidation of images from separate proprietary archives into a VNA can provide immediate benefits associated with economies of scale, and can also provide an environment for improved access to images. This is possible because images stored within the VNA are morphed into a standard DICOM format, thereby enabling access by virtually any DICOM viewer. In some cases, images that have languished in older archives are now revitalized, becoming available to a much larger and wider clinical audience.
Some of today's images are large in size (many in excess of a gigabyte), and the sheer volume of these large data sets is growing. Fortunately, vendors have anticipated this issue and offer life-cycle management programming to automatically address storage retention. Older studies can be readily archived to a less-expensive technology.
A VNA can also deliver the benefit of better performance. Implementation of the VNA across two server configurations will improve performance through automated load balancing, and users can be automatically reassigned if one server should fail. This approach offers availability that rivals fault-tolerant architectures.
With images being stored in a VNA, individual imaging departments are no longer beholden to legacy challenges when shopping for new PACS systems. The daunting task of migrating old images to a new vendor's format is non-existent. The cost and time required for this type of image migration has previously stifled many attempts to acquire newer, more innovative solutions.
With a VNA approach, an organization not only addresses the need to centralize images created within the enterprise, but can also bring in the significant number of studies that originate from outside of the enterprise. Let's take a brief look at how that can be accomplished.
Today, the most popular method of medical image exchange is accomplished via CD. After an image is created and reviewed by a clinician, a copy of the image and the report is burned into a CD. Often, CDs are given to patients, so they can transport it to the next encounter. (Note: The older IT folks among us will recognize this approach as an updated version of the time-honored “sneaker net” from the 1980s.)
Unfortunately, this approach is not efficient or effective. Patients often forget to bring the CD to their next medical encounter. Or the data on the CD is incompatible with the technology at the receiving physician's office, thus it is unable to be loaded and viewed. Even the most capable imaging departments report success rates of less than 90 percent when attempting to view CD images. This inability to view the image often results in treatment delays or additional (redundant) tests being ordered. If the CD being transferred is part of a trauma case, the impact can be significant to the patient.
What can address this challenge? An intelligent DICOM gateway, which serves two purposes: it eliminates the need for a CD to be created, and it enables successful uploading when a CD arrives with the patient.
The optimal scenario is to avoid the creation of a CD in the first place. The gateway would be used to receive an electronic copy of the study via secure Internet (HTTPS). Once received, the gateway would perform appropriate DICOM tag morphing. This would enable the study to be viewed in the native PACS of the receiving site. Additionally, the gateway would send text messages or e-mails to receiving physicians to inform them that studies were available for review. After review, the study would be saved to the VNA. Because this repository would store images from all departments and also be able to import images from outside sources, it would provide a true patient-centric view of medical images.
With a VNA and a DICOM gateway in place, an organization is well on its way to the implementation of a comprehensive enterprise imaging strategy. The third leg of the stool is the implementation of a zero-download, browser-based DICOM referral viewer. This viewer would be able to provide access to diagnostic images from a PC, an iPad or a smartphone, thus giving near-ubiquitous access to any DICOM image as clinicians and patients move between environments — medical office locations, hospital and home.
Great value can be derived from the implementation of a VNA for image storage, along with a DICOM gateway and optimized viewing. This technology and model are used widely in other industries. The deployment of an enterprise imaging management solution is achievable. The question then presents itself: Is it time to go back to the future, to leverage technology and achieve the reality of this vision?
Michael Warthen is responsible for corporate strategy at Merge Healthcare.
For more information on Merge Healthcare solutions