What makes up a medical imaging system?

History of the medical imaging system: PACS origins

So what is medical imaging? To understand that, we need to understand its origins.

Over several decades, medical imaging has progressed from a predominantly analog (film-based) world to a digital one. Major imaging advancements, including CT, MRI and ultrasound, have been key drivers in this evolution. 

As more image information was acquired digitally, the marketplace demanded a means to move that information, centrally store it and display it. For example, consider a facility with two self-contained CT scanners, where these devices acquired, stored, displayed and individually archived studies. Any study acquired on scanner A could only be viewed and stored on scanner A. Any study acquired on scanner B could only be viewed and stored on scanner B. There were no means for images to be sent to a common system for archival and viewing.

It might be said that these early days were the advent of picture archiving and communication systems (PACS). The initial systems utilized commercially available network technology -- Ethernet -- and both online (hard disk drive) and offline (magnetic tape) storage. They also customized image displays -- as commercially available technology did not yet have adequate resolution -- to achieve independence from imaging devices.

The earliest PACS mimicked film alternators, devices for viewing multiple films. They were somewhat limited in their ability to display and manipulate images. Because of varying image manipulation applications and technical requirements, individual workstations evolved around specific applications. 

Today, PACS have achieved a nearly complete saturation among North American-based healthcare providers. Modern-day PACS rely on either a radiology information system (RIS) or an electronic health record (EHR) for patient demographics, as opposed to the stand-alone systems they originally were. Commercial computer technology has progressed to the point where most current-day PACS capitalize on commercial off-the-shelf (COTS) components for processing, displaying and storing images.

Viewing requirements

As PACS have matured, so has the mechanism for handling images, namely the digital image communications (DICOM) standard. Originally conceived as a means for standardizing the communication protocol for radiology images, DICOM has evolved to include extensions for cardiology, mammography, gastroenterology and other medical modalities. Without such a standard, it would be difficult to share image studies between disparate devices, as they would need to interpret multiple image formats.

The need for standardization, as well as the need for high-end graphical processing units, resulted in advanced visualization workstations for such applications as 3D mapping of the brain and spine that were independent of PACS workstations. Vendors and users did not feel the cost of such dedicated hardware within a PACS workstation was necessary. Clinicians needed access to images, but aside from some specialties such as orthopedics and neurosurgeons, requirements were generally simpler than full diagnostic workstations. Even so, they still exceeded general technology in terms of display resolution, so they were typically custom hardware. 

Using COTS technology, today's display processors possess the necessary processing and resolution capabilities to handle most applications. The advent of newer technology such as server-side rendering has enabled broader application of clinical viewers. By sharing a server between multiple users and taking advantage of HTML5 web technology, broader viewing applications are emerging that offer effective performance and security. 

But there has been an expansion in PACS technology beyond radiology, for broader clinical use.

Expansion beyond classic 'ologies'

Radiology and cardiology represent areas of high volume usage of medical imaging technology, but there are other clinical areas with high demand as well. For example, dermatologists frequently use imaging to follow the course of a treatment or procedure. Similarly, surgeons oftentimes document procedures for later reference or for teaching purposes.

Imaging informatics has a role to play in the changing healthcare environment.

While many images are acquired, there are varying reasons for accessing images. Radiologists access them to provide an interpretation for other clinicians. Clinicians might access them to follow treatment.  There may be different requirements for retention. Radiographic studies are kept for extended periods, such as for mammography or environmental cases. Dermatological images may only be important during the period of treatment. This has resulted in a fragmented approach to the storage of patient image information, as individual service lines managed their own information.

This has been disconcerting to IT staff, as they may have to maintain multiple storage environments -- a complex and expensive infrastructure to manage.

Infrastructure rationalization

Expansion beyond the classic multiple image-storing environments represents an opportunity to begin to rationalize infrastructure and upgrade or change medical imaging software. The most frequently addressed area is image storage. These applications, known as vendor neutral archives (VNAs), represent such a capability. VNAs enable images from multiple sources, and potentially in multiple formats, to be stored using a singular application on a common storage infrastructure. 

VNA applications can manage a change in infrastructure that is transparent to clinical staff using the medical imaging system, such as migration to a new storage media. VNAs can manage retention rule sets centrally, thereby relieving individual systems of retention responsibility.

Beyond DICOM

Handling images from multiple service lines may mean accommodating images that are not acquired in DICOM format. One approach has been to encapsulate these images within DICOM, but a new standard has emerged to address this. The Integrating the Healthcare Enterprise initiative has developed a protocol known as cross-document sharing, or XDS. An extension of XDS known as XDS-i is specific to imaging content. The standard has a structure for handling patient demographics, while enabling storage of multiple standard image formats in native form. VNA devices have initiated the deconstruction of the PACS trend as they uncouple image storage and image display from the primary PACS application.

Importance of the cloud medical imaging systems

While VNAs have been effective in centralizing image storage, the sheer volume of imaging data can overwhelm internal IT infrastructures. A promising alternative is storing medical images in the cloud. The cloud can be used for both primary storage, as well as supplemental storage when internal capacity is exceeded. The cloud has the potential to further expand PACS applications by means of software as a service, pushing more functionality into the cloud. 

The primary concern with the cloud has been with security and data ownership. Various authentication methods and secure data centers are working to actively alleviate these concerns.

Relevance in today's healthcare environment

Healthcare reform has accelerated the use of electronic health records (EHRs). EHRs have primarily emphasized transactions and documents, not images. EHRs can document a patient visit or procedure, and can present lab and radiology results.

Recent efforts have enabled EHRs to utilize application program interfaces to directly launch universal PACS viewers to display associated images. This can be done directly, without the need for the clinician to separately access another system such as PACS, and with greater security and protection of protected health information.

Healthcare policy impact

Accompanying healthcare reform, the Centers for Medicare & Medicaid Services (CMS) identified an incentive program for the adoption of EHRs, with defined clinical criteria in progressive stages. As part of this progression, CMS broached the subject of patient enablement by means of patient portals, with incentives to providers to offer electronic patient access to medical records. 

While not a direct requirement, there exists the capability of using imaging informatics technology to deliver images to patients. Technologies such as server-side rendering are ideal for this as they would enable a patient to view images without the physical transfer of imaging data, reducing security risks. VNAs and universal image viewers that can display all image formats are instrumental in helping enable patient access to the image content of their medical records.

Medical imaging informatics' role in healthcare

Imaging informatics has a role to play in the changing healthcare environment. An aspect of healthcare reform is a shift from a fee-for-service approach to value-based care. Population health management is a means for treating patients as part of an entire population, as opposed to on an individual basis. For example, a patient with a lifestyle that would predispose them to diabetes could be identified through population-based analytics and managed with preventative care so as to avoid more expensive treatment of diabetic conditions.

A medical imaging system could have a positive effect on a population health management strategy. One active area of development involves the application of IBM's Watson Health initiative. Watson offers the opportunity to analyze millions of images of heart disease diagnoses, with the potential to more easily spot heart disease from a data set of cases never before seen.