The Case for Reconstructed PACS: It's the functionality that matters

2016 01 27 15 42 41 618 Benjamin Menashe 200

Much has been made of the supposed "deconstruction," if not eventual demise, of the traditional radiology- or cardiology-centric PACS as supplied by a single vendor. Rumors of this demise have been greatly exaggerated, however.

The term Picture Archiving and Communications System implies at least two distinct functions: storage and distribution (including viewing). Additional functions related to workflow, analytics, and reporting are expected. The term, though, refers to a concept and feature set and is not constrained to any particular specialty, architecture, or implementation.1

The industry term "deconstructed PACS" means a standards-based strategy to separate the core components of PACS (archive, viewer, workflow, reporting) and reassemble them with a vendor-neutral approach for enterprise-wide archiving, viewing, workflow, and reporting of images and multimedia objects.

Menashe Benjamin, PhD, of Carestream Health.Menashe Benjamin, PhD, of Carestream Health.

Regardless of what you call it, you need an enterprise imaging platform: a patient-centric, multisite, standards-based platform for data and workflow sharing, reporting, and archiving. The platform must support DICOM and non-DICOM images and multimedia objects (videos, voice files, waveforms, etc.). You need a platform to build on, not a closed black box.

You need images from all "-ologies" to be handled as painlessly as those from radiology and cardiology, allowing for the idiosyncratic workflow associated with their capture. You need a platform that will seamlessly integrate with your electronic medical record/electronic health record (EMR/EHR)2 to trigger image display, and enable it to deliver a complete longitudinal medical record to caregivers, including all medical images (i.e., image-enabling the EHR).

You need a complete solution that addresses all of the features that matter to your organization. Make a list of those features before purchasing. Choose one vendor to supply an entire integrated and tested working solution, or choose to integrate components from different sources and assume some of the risk and support burden yourself (including dealing with finger-pointing among vendors).

Don't purchase one component and only then start thinking about how and what to add to it. Purchase from the fewest number of suppliers necessary to provide a flexible -- but complete -- well-integrated, high-performance (fast), scalable solution.

Delivering higher quality patient care at lower cost has always been an organizational imperative, but never more so than now, with terms like "value-based care" entering the vernacular. Some argue that a paradigm shift is being driven by the current confluence of ubiquitous EHR deployment, consolidation of smaller providers into larger enterprises (requiring multisite image management and workflow), and increasing recognition of the utility of a single local or remote access point to all available information. They suggest that vendor-neutral archives (VNAs) and universal viewers are "disruptive innovations."

Avoiding disruption

As a customer, you might not be as excited about "disruption" in the operation of your own organization. Fortunately, you can respond to the forces of change incrementally.

PACS have evolved to make greater use of commodity hardware and software using open industry standards, and with the fall of communication costs, improved technologies such as Web, virtualization, and portable devices have introduced architectures that previously weren't feasible.

Enterprise-wide imaging and cross-enterprise image sharing are best served by a PACS that has evolved into an enterprise imaging platform, not one that remains a closed product or is a crude amalgam of separately sourced components. The platform you need must allow access by externally supplied components, without sacrificing the performance (speed), integrity, stability, and maintainability of the whole.

There are contemporary single-vendor solutions that offer platform architectures that are built with interoperating components organized in terms of horizontal layers of generic functions (storage and data model, user management, security, auditing, etc.) and vertical application-specific functions (archiving, diagnostic viewing, enterprise viewing, image sharing, reporting, worklist management).

Many of these components are accessible to external systems through standard interfaces, to the extent that they are defined, or well-documented proprietary interfaces, when they are not. This allows any of the application-specific functions to be replaced by a third-party, "best-of-breed" component.

This single-vendor approach allows for a more sophisticated composition of services to provide improved performance and functionality that may be difficult and costly in a deconstructed system from multiple vendors. Two examples of such services are as follows:

  • Multimedia interactive reporting: This allows automatic incorporation into the imaging report of images, structured measurements, hyperlinks from the report text to specific bookmarks on the images, hyperlinks to prior images and reports, and hyperlinks for instant communication between stakeholders. Such reports can significantly improve the quality and productivity of imaging services and patient care.
  • Cloud-based enterprise imaging platform: It would be complex to implement -- and not cost-effective for an integrator to provide -- a solution based on several "best-of-breed" components delivered from the cloud.

It is our contention that this view of PACS as an enterprise imaging platform will serve the customer best.

What you see is what you get

Viewing capability is the most visible and most critical function for user satisfaction and efficacy. Missing functionality or delayed response leads to frustration and dissatisfaction. In particular, specialty viewers are where the expertise and domain knowledge of the vendor is most required.

The "universal viewer" claims to be a potential solution, but with respect to what criteria is a viewer truly "universal"? Is it able to satisfy any user's need for diagnostic viewing or review? Is it able to display any type of image, with any related clinical content? Is it able to interface to any source of images, via any of several competing standards? Is it able to run on any platform, with any security model for access control? Can it do everything without compromising performance, quality, and features?

Too often, universal viewers reflect the intersection rather than the union of all features required, or they perform poorly over standard interfaces without their own dedicated cache, database, and server, and are satisfying to no one.

Serving you best

You as the customer need the option to use best-of-breed components to substitute for or supplement those from the primary platform provider. Components designed and tested by the same vendor will very likely perform better than those separately sourced, but if for some reason you prefer to use a third-party component, you should be able to do so with minimal expense and without sacrificing performance, usability, scalability, and reliability.

Note that data replicated in multiple different databases and archives must be made consistent. Poorly matched, separately sourced components may need their own databases or caches to function and may be difficult to synchronize, raising safety, scalability, and efficiency concerns.

Before you commit to a hypothetically plausible integration, make sure that you see it working in production somewhere in the real world on a scale comparable to yours. Be realistic in your estimate of the total cost of ownership. This cost includes not only initial integration, but also long-term integration support. Separately sourced components will evolve independently, and some may become obsolete or incompatible and need replacement.

Decades of experience should serve as a warning to would-be at-home integrators. PACS integration with RIS, HIS, and reporting systems has never been a trivial undertaking. The current trend toward consolidation of all administrative and clinical functions into a single-vendor supplied EHR (with incorporated HIS and RIS functions) suggests that there is value in a single source for complex systems.

Reconstructed PACS

Independent third-party VNA, viewer, and workflow vendors, each offering only part of the entire solution, have led traditional PACS suppliers to realize the importance of transitioning their products into more flexible platforms with standard interfaces. There is no reason a PACS archive cannot implement all the features of a VNA, for example.

On the other hand, independent component vendors have recognized the risk of having too narrow a product for the majority of the market, and they have begun to acquire, consolidate, integrate, and develop missing components in order to be able to provide a more complete offering.3,4,5 The result looks a lot like a traditional PACS, the only difference being the vendor -- not any purported "neutrality."

Far from PACS becoming extinct, the "reconstruction" of the "deconstructed" PACS has already begun.

All in all, the customer is better off for the industry having gone through this superficially disruptive exercise. Today, there is a much greater recognition by both customers and suppliers of the need for a flexible and extensible, patient-centric, standards-based platform on which an enterprise can build, and with which it can quickly adapt to rapidly evolving needs. Combined with the appropriate EHR, the modern platform will help you achieve the Holy Grail of "all available information accessible from anywhere."

If you must, call it an enterprise imaging platform, when the word PACS conjures up too much baggage or is insufficiently engaging to corporate executives. But ultimately that's what you need, a Picture Archiving and Communication System by any other name: The names of things do not affect what they are.


  1. Huang HK. Short history of PACS. Part 1: USA. Eur J Radiol. 2011;78(2):163-176. doi:10.1016/j.ejrad.2010.05.007
  2. Garrett P, Seidman J. EMR vs. EHR -- What is the Difference? Health IT Buzz. January 4, 2011. Accessed January 27, 2016.
  3. Lexmark acquires Claron Technology. Lexmark website. Published January 5, 2015. Accessed January 27, 2016.
  4. Fujifilm completes acquisition of TeraMedica, Inc. Fujifilm website. Published May 13, 2015. Accessed January 27, 2016.
  5. Karos Health acquires Medical Insight. Karos Health website. Published August 26, 2014. Accessed January 27, 2016.

Menashe Benjamin, PhD, is vice president of healthcare information solutions at Carestream Health.

The comments and observations expressed herein are those of the author and do not necessarily reflect the opinions of

Page 1 of 775
Next Page