We noticed you’re blocking ads

Thanks for visiting CRSTEurope. Our advertisers are important supporters of this site, and content cannot be accessed if ad-blocking software is activated.

In order to avoid adverse performance issues with this site, please white list https://crstodayeurope.com in your ad blocker then refresh this page.

Need help? Click here for instructions.

Today's Practice | May 2013

Strategies for Implementing a Data Management System

Centralizing data into a PACS can improve workflow efficiency.

Bascom Palmer Eye Institute (BPEI), part of the University of Miami Miller School of Medicine, is a large regional multi-subspecialty ophthalmology facility. This tertiary care and teaching center has multiple buildings on its main campus and satellite offices in surrounding communities. Currently, physicians at BPEI treat approximately 250,000 patients and perform more than 12,000 surgeries annually.

This article details my experience implementing a data management system at a high-volume institution and the benefits such a platform affords a practice and its physicians.


BPEI became interested in the evolving technology of digital photography in the early 1990s, when the first ophthalmic digital cameras came onto the market. As a teaching institution, it is incumbent upon BPEI to become familiar with every device available. Many ophthalmology departments or private practices may have single-vendor solutions or a limited array of imaging systems, but, at BPEI, we tend to have multiples of every device. For a decade, the institute experimented with different vendor imaging systems and by 2002 had acquired a significant number of devices and clinical imaging technologies for a variety of subspecialties including cornea, retina, glaucoma, pediatrics, and neurology.

By 2004, there was a pressing need for us to start gathering all of our images and data into a single source. We made the leap to an integrated picture archiving and communication system (PACS) from Ophthalmic Imaging Systems (OIS), now known as the Merge Eye Care PACS from Merge Healthcare. At that time, the digital imaging and communications in medicine (DICOM) format used widely in radiology was mostly unknown in ophthalmologyóor perhaps, more accurately, not supported among ophthalmology imaging vendors. In the intervening years, more ophthalmology vendors have become DICOM-compliant, and there has been an evolution toward exporting imagery or reports in a meaningful manner. One significant change was that vendors began offering options other than DICOM (while ophthalmology-based DICOM standards were being hammered out), such as offering pairs of images and XML data files together, so that users could extrapolate the data associated with each report. Optical character recognition technology also became more reliable and flexible in the manner in which data could be exported.

Another major change was the ability to centralize the process of importing data to the system. When we began this consolidation project, we found many of our instruments were not initially network-compatible, so there were small non-networked computer workstations for each imaging modality scattered all over the building. As these third-party systems evolved and became networkcompatible, this allowed us to move the processing capability into a centralized, more secure and stable format.

We have continued to expand the input systems and instruments that we bring in to the Merge platform. Within the past 4 years, the University of Miami, along with many other large institutions, has taken on the task of implementing an enterprise-level electronic medical records (EMR) system. In anticipation of moving toward a paperless environment, we not only needed to bring everything into the digital format but also had to prepare for the increased load on our imaging distribution system this would entail. BPEI has been averaging 5,500 imaging exams per month for more than 2 years in our main Miami office alone. That is not 5,500 images but rather 5,500 imaging examinations, many of which may include hundreds of images each. We have seen significant challenges in handling this volume.


By 2010, our initial investment of system hardware including servers and storage had reached the end of its serviceable life and were due for refreshing, and, coincidentally, our information technology (IT) department was making a push toward virtualizing much of its own infrastructure. Our IT staff proposed the concept of virtualizing our Merge Eye Care PACS, and together with Mergeís technical staff we began the process. Virtualization allows a server, or an operating system, to run on top of the same hardware in parallel. For example, four or five server operating systems might all be running in a Windows (Microsoft) environment, each separate from the other, but all running on the same physical hardware. In examining lessons learned over the past several years, I felt that there was a distinct advantage to being able to divide what we do into smaller chunks. In a traditional server environment, it is routine to have multiple system processes (eg, importing, indexing, and database management) running in the same environment. By separating these processes, if a problem were to occur, it could be more quickly identified and isolated. Application and operating system updates or any form of troubleshooting becomes more manageable, with considerably less risk of system downtime.


The impact of moving away from the paper chart and having the ability to access patient images and data anywhere and anytime has been tremendous. Although we have three satellite facilities, they are not far apart, and given the population density in South Florida it is not uncommon for a patient to be seen in one office at one visit and a different office the next. Having access to that patientís full history and current imaging at all locations is extremely useful. With a data management system in place, the days of delayed access to paper charts are virtually over. (Pun intended.) A BPEI physician, who was routinely seeing 95 patients per day, makes the point that if he could save 1 minute for every patient he saw, that would amount to 95 minutes out of his day. The time saved looking up patient information or imaging exams could be better used to see new patients or spend more time with existing patients. Time is of the essence, and having this ability is important to clinicians.

Transitioning to a paperless environment can also be associated with potential cost savings including costs for paper, printer ribbons, printer services, toner cartridges, and chart storage. Several years ago, at our Miami location, we outgrew our onsite capacity to store our paper charts, which necessitated contracting offsite storage. Being able to relieve that burden and put all records in a digital format saves money, time, energy, and staff that we would pay for as long as we continued to use a paper method.


Server virtualization is advisable and worth the effort. IT departments would love the idea of moving in this direction.

The Merge Eye Care PACS is geared toward the clinician. It presents the physician with a number of tools, including creating image overlays, tracking patientsí progression over time, using an auto-scrolling format, and condensing imagery into thumbnails. We have had nothing but positive responses from the clinicians on how user-friendly and productive it has been for them.

The Merge Eye Care PACS platform also provides enhanced security. There is the ability to separate administrators from general clinicians, so that some documents are in a read-only, view-only environment to eliminate the danger of malicious damage. The system interfaces directly with active directory environments, and all passwords and user information can be tied to that.

In terms of performance flexibility, when a user logs into the system, it measures the available bandwidth. The system can determine when the user is using a lowbandwidth connection, such as a slower DSL connection, and can preprocess images and tailor its response to the available speed, providing images in the quickest manner. If the user wishes to see full, uncompressed images, that is an option as well. All three of our locations have different bandwidths, and it is important for us to have the same performance level at the satellite offices despite this.

For institutions considering implementation of a data management system, I strongly suggest that the people involved in this decision ensure the data integrity of a device before that data is merged with the centralized system. It is advisable to exhaust all capabilities of third-party systems to help benefit the data management system as a whole. Finally, the support staff of the PACS manufacturer should be a helpful and valuable resource. The staff should be willing to work with vendors of the individual imaging technologies to adapt the system to the practiceís individual needs. Issues with vendors can take months to resolve, but the data management system manufacturer should work to resolve issues immediately.

Richard Stratton is the Senior Systems Analyst at Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, in Florida. Mr. Stratton states that he has no financial interest in the products or companies mentioned. He may be reached at e-mail: rsstratton@med.miami.edu.