Conventional wisdom less than a decade ago held that digital radiography – at the time still an emerging technology – would prove transcendent and cause its chief rival, computed radiography, to quickly go the way of the buggy whip. Alas, as the years passed, it became obvious that such was not to be the case.

“DR arrived with the promise of being able to dramatically increase radiography throughput in comparison to what was possible for CR, and it was chiefly on that basis that the disappearance of CR was predicted,” says J. Anthony Seibert, PhD, professor of radiology at the University of California, Davis.

Both CR and DR output a digital x-ray image rather than traditional analog film. But the way the digital image is acquired can give DR a slight edge over CR. In a nutshell, DR couples a flat-panel, thin-film transistor (TFT) array or a charge-coupled device (CCD) array with an x-ray absorber to capture the produced image on a built-in phosphor or photoconductor plate. The captured image then is converted internally by the DR technology to digital form and distributed to a nearby workstation for technologist review and handling.

CR works in a similar fashion, except that the capture plate is not built in. It is contained within a cassette that the operator removes and hand-carries to a processor that might be installed in the same room as the CR unit or, more likely, down the hall in a centralized location for use by other CR technologists in proximity.

“The need to transport the cassette to a reader makes CR a labor-intensive modality, very much like analog screen-film x-ray imaging,” says Seibert.

DR systems also offer better detective quantum efficiency than CR, Seibert points out. However, both modalities feature wide latitude and increased dynamic range that helps sharply reduce the need to repeat procedures caused by either overexposure or underexposure.


Nevertheless, there are certain advantages CR enjoys over DR. Lower up-front cost is one. Versatility is another.

“Where CR has a leg up on DR is in positioning flexibility for many types of examinations,” Seibert says. “Also, the detectors are less fragile; it’s harder to damage them if manhandled or even droppedwhich helps explain why CR is well suited for portable imaging and hospital bedside radiography.”

That sums up the impression of CR held by Lorraine D. Kelly, RT(R), clinical delivery process manager for Lahey Clinic Hospital in the Massachusetts cities of Burlington, Peabody, and Lexington. Her enterprise – producing about 330,000 studies a year – offers diagnostic, interventional, and general radiology. Among the modalities used are both CR and DR.

“We have more than 30 CR machines but only one DR unit,” she says. “Cassetteless CR offers the best of both world: image quality and increased patient flow.”

Kelly prefers CR to DR on a number of grounds, one of which being that implementation of CR is less expensive and complex. She also likes CR for its go-anywhere and shoot-at-odd-angles capabilities. Image quality represents yet another plus.

“In our experience, CR image quality on large body parts such as chest and abdomen has been better and more consistent than DR,” she says, while conceding that DR image quality is excellent on small body parts, extremities in particular. “Also, technologists acquire the CR images in a fashion that is similar to how it’s done with film – they feel they can be more productive with CR because they move things in ways they’re already familiar with.

Techs prefer cassetteless CR for outpatient walking-talking exams but prefer cassette-based CR for more complex images. We have been unable to satisfy our radiologists’ requirements for chest and abdomen on the DR unit.”

Even so, Kelly does not consider CR to be a perfect technology. “CR has the same limitations as film,” she says. “It’s sold as having advantages over film, which it doesnamely, a small window of repair for the image and huge benefits relating to filmlessness. But it really has similar issues as film in regard to the physics of radiation. When doing lateral c-spines or shoot-through c-spines for trauma, if you can’t penetrate the part effectively, you can’t window that image into being a perfect image.”

The main reason why her enterprise wanted to explore DR was to improve workflow while increasing image quality.

“The biggest selling point for DR was workflowno cassettes to run to the reader; you could take the images faster and save time,” Kelly offers.

However, the DR technology presented some challenges. “The first DR problem we had was with positioning,” she says. “Effective penetration or effective exposure to the array with large body parts was very inconsistent. As a result, our repeat rate increased. Theoretically, repeat rate goes down with DR because you have the ability to manipulate the image. But in reality, you don’t have as much room for error as you’d think. And if you’re outside of the window of error or outside of the range of exposure, there’s nothing you can do but repeat the imageyou can try to manipulate until you get a decent-looking image, but it will be one that’s missing data, which puts you at risk of having an image unable to reveal everything relevant.”


While Lahey Clinic Hospital may have found DR a less-than-ideal solution, the same cannot be said for Kaiser-Permanente Hospital in South Sacramento, Calif.

“We use CR and we like it, but when I was given the opportunity to oversee conversion of the radiology department to digital, I decided to build as many DR rooms as I could,” says Ruth Dwight, RT, director of radiology.

Having now done that, Dwight has few complaints. “DR has been a good choice for us,” she asserts. “Because of DR, we no longer have the repeat rate we once did. It’s now very, very lowwith the exception of where the positioning is wrong to begin with or where we have instances of the patient moving around on the table. We’ve found that DR gives us a good window in which we can enhance the image density and contrast.”

The lower repeat rate means Dwight’s technologists are more productive. But even if they were repeating at the higher rates of times past, they would still be more productive than if they were imaging with CR, Dwight insists, simply because of DR’s no-cassette processing.

“Initially, I thought I’d see a 35% increase in throughput in my daily sections,” she says. “To my surprise, throughput was up by 65%. In this enterprise, that kind of throughput gain makes a huge difference.”

Dwight asserts that DR has also helped stem the incidence of technologist injury. “My workers’ compensation injury rate has gone down more than 60% since converting to DR,” she says. “Without those cassettes to carry, shoulder and back injuries are few. And because the injury rate is down, productivity stays up.”

Dwight recalls even being thanked for that from an unexpected source. “At the end of a job-performance review with one of my technologists, I asked him if he’d like to offer input that would help me better support him in the future,” she recounts. “He declined that opportunity but did take a moment to tell me how much his wife appreciated our move to DR. His wife doesn’t work here, so I asked him to explain why she’d be grateful. He replied, Because now, when I come home at the end of my shift, I’m not exhausted as I was when I was working with CR; I come home and I have energy, and my wife really likes that.'”


With pluses and minuses evident in DR and CR alike, some decision-makers wonder which of the two has the most sustainable clinical future.

For a partial insight, comments Mark D. Perna, RT(R), PACS administrator for St Luke’s Hospital and Health Network in Bethlehem, Pa, which consists of four medical centers (three already linked by PACS, and plans for the fourth are in their infancy) and a 310,000-study-per-year radiology piece that includes 13 CR units and three DR rooms: “If you’re setting up a filmless environment as we have here, your film-based x-ray processes are going to have to be replaced before PACS can be introduced. As radiography equipment needs to be replaced because of either age or failure, the question you face at that point is, what will you replace film x-ray with? The choice comes down to CR or DR. I think most enterprises will look first to replace with DR, especially if they’re moving to PACS.”

That is how DR came to be at St Luke’s. However, soon after acquiring DR, the network also ventured into CR as part of a PACS rollout.

“Our thinking was that, if money were no object, we’d replace all our rooms with DR and maybe have a few CRs to handle portable requirements,” he says. “But money was an object, which is why we also began implementing CR. We recognized CR as a comparatively economical choice that installs quickly and allows you to get up and running quickly, without a disruption of service. One day you’re using film, the next you’re using CR. By contrast, with DR you replace film by first having to tear out an x-ray roomthat makes for a time-consuming and task-intensive installation.” Perna maintains that there is a place now and in the future for both CR and DR.

Seibert, the UC Davis physicist, is of the same mind and appears convinced the two technologies will each become more appealing in the years ahead.

“A number of innovations are probably going to occur in DR,” he says. “For example, we’ll be seeing continued improvements in DR phosphor and photoconductor technology, which will lead to increased readout speed. That’s also going to make dual-energy and tomosynthesis a practicality. Farther out in the future, we might see those relatively fragile glass TFT arrays replaced by much hardier yet flexible substratesthese will allow DR detectors to be more robust, which would be highly desirable in bedside applications.

“As to CR, manufacturers already have introduced a thicker phosphor layered on a light-transparent base. This enhancement is allowing dual-side readout by incorporating two light-guide assemblies in the reader, which in turn permits capture of a greater fraction of the light signal with the use of a smaller laser beam. By some measures, this causes detective quantum efficiency to rise by about 50%.

The Big DR Bang

In January 2002, the radiology department at Kaiser-Permanente Hospital in South Sacramento, Calif, embarked on a daring move-overnight the hospital and its quartet of medical office buildings went from film to filmless.

To make possible such a sudden shift, the hospital installed a PACS, but also acquired and deployed new digital modalities en masse, including five digital radiography and three computed radiography units.

“If I had it to do over again, I wouldn’t do a thing differently,” says Ruth Dwight, RT, director of radiology, whose department is responsible for the production of some 200,000 imaging studies annually. “Prior to coming to Kaiser, I worked in a university setting where there came a time that we discontinued film. The university’s conversion plan called for, among other things, wide deployment of CR. However, at the university, implementation of CR occurred in baby steps. It was thought that this go-slow approach would allow us to avoid a lot of pain. It didn’t. In fact, all that happened was it made the pain last longer than it might have otherwise.

“So, here at Kaiser, in our conversion to filmless, with the implementation of both CR and DR and all the other modalities, I figured that, if we were going to have pain no matter what, why not just do it in a way that gets it over with in the shortest possible time?”

Dwight’s experience at her old place of employment convinced her she could expect at least 6 months of misery by converting in small stages, but would endure no more than 6 weeks of it with an all-at-once changeover. To her delight, the pain in actuality lasted but a single week.

“We planned carefully for it, that was one of the keys,” she says. “Another big part of the success was having made good choices in equipment.”

“Dual-energy imaging, while not new, is likely to become more prevalent due to innovations in both CR and DR detectors, with improved DQE for CR and higher speed readout for DR dual-energy implementations. Previous implementations required either or both multiple exposures and handling of multiple plates, as well as careful image registration.”


These and other advances in both DR and CR, ironically enough, may ultimately render moot the debate over whether it is better to outfit with one or the other. That is because each step forward is causing DR to exhibit the characteristics of CR, and CR to behave more like DR.

“There’s no question that the distinctions between CR and DR are blurring,” says Seibert. “Some DR manufacturers are coming out with portable units. These look intriguing, although they’re not yet really proven. Meanwhile, some CR manufacturers are offering systems with single-exposure, cassetteless dual-energy imaging through the use of automated plate handling, built-in metal filters, and software that speeds image registration enough to allow approximately 40 exposures per hour to be acquired.”

Seibert is reasonably confident this merging of the two will compel radiology enterprises to abandon use of the terms CR and DR in favor of something else.

“After all, these technologies are inherently digital and computed radiography should be considered as cassette-based versus cassette-less in the new digital era,” he says. “In addition, since physicists think about how the x-rays are converted into a signal, perhaps we need to start seeing the differences in radiography technology in terms of the nomenclature of indirect digital capture versus direct digital capture, which is really about the only distinction that might in the end be left.”

Rich Smith is a contributing writer for Decisions in Axis Imaging News.