In November 1998, the Brooke Army Medical Center in Fort Sam Houston, TX, performed what it calls the first commercial upgrade of a conventional analog x-ray table, installing a flat-panel digital x-ray detector in a system that previously had been outfitted with computed radiography technology.
One year and nearly 2,000 examinations after making the switch, the center has found the flat-panel digital images to be generally better than those produced by computed radiography systems. In some ways, however, the flat-panel system is less practical than the CR technology it replaced.
Interest in flat-panel digital x-ray has grown over the past few years, thanks to the development of digital detectors based on amorphous silicon or selenium. The systems are seen by many industry watchers as an improvement over CR because they enable facilities to move images directly into a PACS, rather than requiring a separate reader for digital conversion, as is done with CR.
Thus far, however, most vendors have been concentrating on offering flat-panel digital x-ray as part of new x-ray systems. This results in higher costs for facilities looking to convert their x-ray systems to digital. Going the retrofit route could be a cost-effective option for many facilities that don't want to buy new systems.
To retrofit its table, Brooke Army Medical Center used a flat-panel amorphous selenium detector manufactured by Direct Radiography Corp. of Newark, DE. DRC is a subsidiary of Waltham, MA-based Hologic. The CR images it used for comparison were produced on the center's five computed radiography units, manufactured by Fuji Medical Systems USA of Stamford, CT.
In order to assess contrast and spatial resolution of the flat-panel digital images, images of various sizes and densities from 12 phantom disks were created on each system. They were evaluated by five readers, who were blinded to whether the images were collected with a flat-panel digital or CR system.
Overall, significantly more correct responses were observed with the flat-panel images as compared to the CR images. A second study of 50 clinical images, designed to measure examination times, found no significant difference between the two systems, a phenomenon the researchers attributed to the extra time required to interface with the new system.
The researchers found significantly better contrast and spatial resolution in the direct digital images, and, in a few cases, improved visualization of pathology. Overall, they found the images to be at least equivalent to, and often better than, images produced on the CR system.
Scaling the learning curve
Now that Brooke's flat-panel system has been in place for awhile, and the learning curve has presumably flattened a bit, the center has begun to test the new system more broadly. Specifically, 400 patients and about 1,000 images are now being studied to assess the system's effects on overall image quality, diagnostic yield, and throughput. Preliminary results so far are encouraging.
"Qualitatively, the images are gorgeous," said Dr. Brian Penrod, a musculoskeletal radiologist at the center who has been closely involved with the conversion. However, he cautions that the CR images being used for comparison "may not represent the state of the art in computed radiography."
As for productivity, he says, "Initially the throughput wasn't significantly better, but we're starting to see a turnaround as people get more familiar with the system. We're anticipating an improved overall throughput with the system."
Still, any radiologist knows there are other factors to consider besides image quality, among them the overall practicality of the system. Penrod said DR's lack of portability is the main reason why he could not convert the center's other nine radiography rooms to flat-panel digital.
"Portable exams are still going to be a big issue because that's a good portion of the workload in a department, especially in the hospital," he said.
Penrod believes that deploying nothing but flat-panel systems might be an option in an outpatient clinic where most of the patients are ambulatory. Beyond the portability issue, he said, current flat-panel digital systems can't be used for cross-table lateral work, scoliosis exams, and a few other applications. And they remain more expensive than CR systems.
Judging from what he saw at this month's RSNA meeting, manufacturers are in a race to improve the technology of the both CR and DR systems, with CR manufacturers working to improve image quality and DR developers aiming for more practicality. Who wins the race, he said, is anyone's guess.
"It will be interesting to see how this all plays out in the next couple of years," Penrod said. "It's going to be a little unsettled for awhile."
By Eric BarnesAuntMinnie.com staff writer
December 22, 1999
Copyright © 1999 AuntMinnie.com
