Paul Keough

It’s been argued that repeat/reject analysis, a quality-assurance tool to observe trends and make improvements that help reduce the need for repeat images, is not necessary when using digital equipment. While it is true that digital imaging reduces some error rates, the results on the Mayo Clinic digital poster exhibit at the RSNA meeting showed that mistakes still occur in the digital environment. 

  Jill Lucas, a radiology technologist at the Mayo Clinic in Rochester, MN, presented a poster showing repeat/reject analysis findings collected by a digital radiology information system (RIS) with an automated entry method that allowed technologists to enter the reason for a repeat image by scanning a barcode. The barcode makes it easy for the technician to select a reason for the repeat and provides a consistent method of tracking the reasons. 

  One challenge in the study was that the using the barcode entry method was not mandatory. The technologist was neither rewarded for selecting nor penalized for failing to select a barcode reason for repeat/reject images. 

  However, Mayo Clinic has the ability to check each device and see if the reasons for repeat/reject entered match the data collected in the RIS. Also, visual reminders are available for the technologist. 

  The results of the study show that error rates due to operator error are about the same after introducing digital imaging as before its introduction at the institution. While digital imaging has reduced patient exposure to x-ray dose and lowered the number of positioning errors, the findings still showed that some errors occur at the same rate as with film imaging. 

  “The majority of the repeat images are due to clipped errors, where part of the anatomy is missing, and rotation errors, where the part of the anatomy needed is not lined up,” Lucas said. 

  The Mayo Clinic has guidelines for what percentage of repeat/reject studies is permissible. In areas where the guidelines are not met, attempts are made to correct errors. Quality control in areas with higher rates of repeat/reject digital imaging errors could lead to training sessions to show technologists the proper positioning of, for example, a knee.

Articles

It’s been argued that repeat/reject analysis, a quality-assurance tool to observe trends and make improvements that help reduce the need for repeat images, is not necessary when using digital equipment. While it is true that digital imaging reduces some error rates, the results on the Mayo Clinic digital poster exhibit at the RSNA meeting showed that mistakes still occur in the digital environment.

Repeat/reject analysis makes case for digital training

The artifacts observed in images derived from computed radiography come from a range of sources and user errors. A radiography team from Manipal, India, presented a digital exhibit at the RSNA meeting to illustrate sources of such artifacts and user solutions to overcome them.

Image acquisition artifacts arise from a variety of conditions. Out of a number of cases presented, three case studies illustrate some of them.

In the first case, a Double J stent has been inserted. However, the postprocedure kidney urinary bladder radiograph showed two stents (Figure 1A). The twin, or double, effect occurred because two consecutive exposures were taken during one acquisition.

The key to eliminating this doubling artifact is proper exposure technique. The radiologist should always cross check with the surgeon about the placement of Double J stents when more than one stent is seen on the same side in a radiograph. Another way to spot artifacts in this radiograph is by noting the subtle doubling of adjacent structures, which may not be very evident, as in this example.

Once the radiograph was repeated, only a single stent was found (Figure 1B). “The remedy is proper exposure and proper use of the equipment, which are needed to prevent such misadventures,” said Ashita Barthur, MBBS, of the Kasturba Medical College and Hospital in Manipal.

The second case is a radiograph of the pelvis, which shows an increased density over the sacrum and right half of the pelvic bone (Figure 2A).

The problem here is an uncollimated primary beam that results in an increase in radiation exposure and an image that is not sharp. If the image collimation is not parallel to the imaging plate, proper borders will not be recognized, resulting in a suboptimal image.

The remedy involves proper collimation to match cassette size and body part to be imaged. In the retaken radiograph of the pelvis, details of the sacrum and pelvis are clear (Figure 2B).

The third case shows a radiograph of the chest and upper abdomen. Notice the artificial increase in density over the center of the abdomen at the bottom of the image (Figure 3).

“This [artifact] basically resembles a light bulb with the center of the light bulb at the center of the abdomen, which results in an artificial darkening of the periphery, which can be misdiagnosed as pneumoperitoneum, Barthur said.

This artifact results from using a high-kVp technique in obese patients and improper collimation. The solution to the light bulb artifact is reducing backscatter by lowering the kVp or by more precise collimation.

Poster shows computed radiography artifacts and their solutions

A maneuverable digital radiography system has been selected over a computed radiography system by the emergency room at a U.K. facility. Steven McDonald, director of radiology in the emergency room at Royal Liverpool Hospital gave the rationale and implications for this decision in a poster presentation at RSNA 2009. 

  “When we started, we had just come to a CR solution,” McDonald said. 

  One of the hurdles to overcome was that DR is sometimes viewed as not flexible enough for the emergency room and is not to obtain the same images one could image with a CR plate. 

  To discover if flexibility was possible with DR, McDonald’s team visited a group in Charleston, SC, which made DR in the ER look smooth. 

  “Patients came in quietly; no one was breaking into a sweat. We were impressed that it was not only a DR system,” McDonald said. “but it also had the capabilities to do what a CR cassette could do in terms of flexibility.” 

  In this DR system, a laser line properly orients the limb and the scanner’s flexible head. 



This maneuverability made it the definitive choice for the ER team at the Royal Liverpool and Broad Green Hospital. This DR system has been used in the city trauma center for the last 18 months with impressive time-saving results. 

  The adoption of DR led to a two-minute per-patient time savings over CR. The savings comes because DR does not have to use the cassette, and, if a change is needed, retakes can be obtained much more quickly with the DR system. 

  Finally, McDonald said that the patients are really in tune with the DR. Patients who had been visiting for three years saw big changes from CR to DR. Now, patients wait only a short time before their image is taken, and leave promptly. 

  Said one patient, “It’s like Star Wars.”

A maneuverable digital radiography system has been selected over a computed radiography system by the emergency room at a U.K. facility. Steven McDonald, director of radiology in the emergency room at Royal Liverpool Hospital gave the rationale and implications for this decision in a poster presentation at RSNA 2009.

Digital radiography wins over emergency room in U.K. setting

Optimization strategies described in an RSNA poster presentation from Switzerland were found to reduce total dose exposure by 63%. 

  Developed by the Institute of Diagnostic, Interventional and Pediatric Radiology at the University of Bern, the “EasyDose” system, which is at the center of the facility’s DoseWatchers program, can collect data from any radiation source and manage the exposure parameters. The RSNA presentation was developed by Dr. Rainer Wolf, who inherited the project as his team began monitoring radiation dose with a data retrieval system. 

  Wolf used the retrieved information to automatically add more than 62,000 pediatric patients into the database, allowing a prospectively designed two-year dose optimization study to reduce the radiation burden for children. During the first year, 14,145 images of 7462 patients were collected; in the second year 14,726 images taken of 7928 patients were gathered. Distribution of CR/DR studies was about 80% DR and 20% CR in the first year. 

  Optimizing dose then involved three steps: copper filtration, kV/mAs reduction according to image noise, and bilateral lead collimation in anteroposterior views of the spine. In the first step, a copper filtration reduced the effective dose from 0.234 mSv to 0.14 mSv. The second step reduced kV/mAs to 0.039 mSv. A third step in full-spine imaging involved use of bilateral lead collimation to block peripheral organs (breast tissue) and reduced radiation to 0.035 mSv. 

  Spine imaging represented only 3% of all studies (Fig 1.1), to but 38% of the dose load by dose area product (DAP) (Fig 1.2). 

  “This was the most dramatic difference and became the first item to attempt to resolve,” Wolf said.  



  The spine is a large area for exposure, so it represents a relatively larger dose than other body parts. But after two years of the dose optimization program, not only was the total DAP 63% lower, the relative DAP to the spine was reduced from 38% to 19% (Fig. 1.3). 

  The lower dose studies were then measured with a phantom that computed image quality for contrast and detail. The results showed the detail value remained the same for all optimized doses and only a small amount of contrast was lost. 

  Pediatricians who requested the studies were not informed about the dose reductions and had no complaints, with the exception of one orthopedic surgeon who wondered why some unstudied anatomy was cut off. However, when it was explained that lead collimation was used as part of the dose reduction program (and as breast cancer prophylaxis), the physician was pleased. 

  “Digital imaging allows lower dosing that reduces the radiation burden significantly for [the] pediatric age group, and also would benefit adult patients,” Wolf said. 

  The results presented were for mixed CR and DR. An even greater effect should be observed in DR alone, with perhaps half the dose used for DR compared with CR.

Optimization strategies described in an RSNA poster presentation from Switzerland were found to reduce total dose exposure by 63%.

Author | Paul Keough

Articles

New content coming soon, please check back later.