Although much improved in recent years, high-resolution monitors used for primary diagnosis still have relatively short life spans and need frequent service. As a result, routine maintenance and quality control are critical for these workhorses of the filmless radiology department.
To assess image quality and monitor failure rates at their workstations, investigators from the University of Maryland School of Medicine and the VA Maryland Health Care System in Baltimore, MD, reviewed the maintenance records for their high-resolution monitors (2048 x 1536 pixel) and randomly tested their image quality. The team, led by Dr. Eliot Siegel, described the experience in the August Journal of Digital Imaging (Journal of Digital Imaging, Vol. 13, No. 3 (August), 2000: pp114-118).
The researchers looked at the vendor’s maintenance records over 17 months, during which time 20 of the institution’s 24 high-resolution monitors suffered major hardware failure. This represented a failure rate of approximately 5% per month with an average time until a major repair of 1.7 years, according to the researchers.
"No picture" was the most common reason for a monitor service call (35%), followed by "low luminescence" (22%), "poor image quality" (17%), and "distorted picture" (9%). Other miscellaneous causes (totaling 17%) included "bad smell," "bad switch," "lines projecting over image," and "partial display of image."
In 61% of the service calls, the PACS vendor had to replace the monitor, while 22% of calls required replacement of the monitor's video card. The rest of the repairs (17%) included miscellaneous causes such as a simple reboot of the workstation, repair of the monitor switch, and/or modifications of the control panel settings.
"Given the potential loss of productivity with failure of a video card, this relatively high failure rate would seem to mandate a policy in which spare video boards are stored on site," the authors noted.
The researchers also evaluated monitor quality through random surveys using a Society of Motion Picture and Television Engineers (SMPTE) test pattern to determine the contrast and spatial resolution of the monitors. Twelve of the 22 monitors were found to be of suboptimal quality, having at least one region in which image quality was judged suboptimal or unacceptable. Of the 22 monitors, 27% had at least one region that was scored as unacceptable.
The clinical impact of monitor quality problems is uncertain, although the researchers have gathered preliminary data for additional studies that suggest a "surprisingly small effect on diagnostic efficacy despite obvious image quality problems shown by review of the SMPTE pattern." The researchers have also generated preliminary date finding that, in contrast to the utility of an SMPTE pattern, radiologists don’t fare well using diagnostic images to determine monitor performance.
"These findings suggest that a technician, technologist, radiologist, or other healthcare worker should utilize a test image such as a SMTPE pattern routinely rather than depend on a perceived degradation in image quality, which has been to shown to perform poorly in the evaluation of monitor quality," the authors wrote. "Given the high prevalence of image problems, it is likely that this assessment should be performed on a regular basis such as daily or weekly rather than on a quarterly basis as suggested by the PACS vendor."
The substantial failure rate of high-resolution monitors should be accounted for in the PACS purchasing process, according to the authors.
"Requests for proposals (RFPs) for PACS and service contracts must specify carefully requirements for monitor image quality and conditions under which the vendor is required to replace these monitors," they wrote.
Overall, the average monitor has an average life expectancy of 2.4 years with an average time to major repair of 1.7 years, an improvement over early PACS monitors, according to the researchers.
"However, these figures remain well below the anticipated life expectancy of other PACS components, including the workstation itself and the PACS server and network, which we estimate will have a useful life expectancy of approximately 5 to 10 years," they wrote.
By Erik L. RidleyAuntMinnie.com staff writer
September 19, 2000
Click here to view all active PACS Digital Community discussions
Copyright © 2000 AuntMinnie.com
