In a presentation at the 2008 SIIM conference, Dr. Paul Nagy, from the University of Maryland School of Medicine, said he considered PACS to be a great system for creating a quality assurance program. This would provide radiology departments a quantitative evaluation of image quality and, in this particular incidence, radiological technologists would be placed under the spotlight.

"Peer review is a powerful motivator that we do not utilize enough," says Nagy. In today's radiology practices most track the level of image quality indirectly by monitoring the RT's rework or better yet the radiologists' complaints. Though useful in trying to minimize errors, voluntary incident reporting does not help raise the culture of quality, he adds.

Utilizing context based integration with a PACS system in the evaluation of a radiological technologist, Dr. Nagy and colleagues created a framework to measure the level of image quality of performance that provided a means for continuous quality improvement and excellence. To do this, researchers distributed adjudicated sampled peer review methodology to the technologists so that they could review the quality of their peers' work. To ensure fairness and consistency in review methods, the technologists' supervisor evaluated all reviews.

At the beginning of each week, the system queried a report repository via a simple object access protocol and randomly selected approximately five per cent of the previous week's studies for review. Each of the six volunteer technologists reviewed 25 studies per week. In addition, supervisors received a similar worklist of the reviews needing approval. Out of the 3,049 studies reviewed, there were 2,738 (89.8 per cent) approved and 311 (10.2 per cent) that were not accepted by the supervisors.

"This tool was found to be unexpectedly useful and provided significant benefits to the imaging department," says Nagy. Researchers and imaging supervisors noted that this data can be used in performance reviews to create a more "in-depth" report card of areas in which the technologists excel, as well as where there is a need for improvement.

Just because a technologist knows what to do, it doesn't mean that they are going to do it. When the time comes for a performance review, "which one are you going to hire or promote? The one who doesn't know what to do but shows room for growth and excellence or the one that has been there for 25 plus years and knows what to do, but doesn't do it?" asks Nagy.

But how do we correlate such a success, you ask? "Look at your radiologist's complaints," suggests Nagy, "have they slowed down?"

Quality assurance plays an integral role in all aspects of radiological imaging, but what exactly is 'image quality'? In an effort to research the exact definition, researchers queried the term online via one of the most respected medical dictionaries available. However, they didn't get very far with the reply: "Sorry, but the Stedman's Online Medical Dictionary doesn't recognize this term".

"The destination of quality can be elusive" says Dr. Eliot Siegel, professor and vice chairman of the University of Maryland Department of Diagnostic Radiology and Chief of Imaging VA Maryland Healthcare System.

In spite of the significant dependence on digital radiography for medical diagnosis and treatment planning, there are no quality assurance standards. With the continued pressure of maximizing productivity and workflow, a growing number of 'images of lesser quality' are often accepted by both technologists and radiologists alike, resulting in the submission of equivocal reports and unnecessary follow up imaging recommendations.

Reject analysis is no longer as reliable as it once was for conventional film screen radiography. Interestingly, Eliot states, "Another overlooked consideration affecting digital radiography QA is the ability for technologists to easily delete poor images without any permanent record." Unfortunately, with little or no research being performed, minimal industry effort is made towards the development of QA enabling technologies.

In creating a multi-institutional digital radiography quality assurance database for education, research and technology development, Dr. Siegel and colleagues also hope to facilitate and develop reproducible and quantifiable QA standards, as well as improving quality and safety throughout medical practice.

After the installation of an automated reject tracking program on 20 computed radiography systems, researchers created a multi-institutional digital radiography database. The reject tracking program required all technologists to enter a cause for rejection each time an image was repeated. Researchers noted, due to the lack of standardized QA terminology, the recorded cause of rejections differed between institutions and technologist, requiring a systematic review form and independent review board.

Including the lack of synonymous QA terminology, the need for a standard QA lexicon and the lack of supporting QA technology, the QA initiative made many interesting observations and created several additional opportunities. Researchers anticipate the QA database will serve as an important resource for the development of a standardized QA lexicon and quantifiable QA metrics.

So, is the beauty of radiological images in the eye of the beholder or the radiologist? It looks like we will need to build an image database strong enough to objectify it.