March 2012

Practice Makes…

By David Yeager
Radiology Today
Vol. 13 No. 3 P. 32

…OK, not perfect, but simulation-based training can improve radiology instruction.

Progress often occurs when someone puts a new spin on an old idea. For example, cases of medical simulation can be traced to the ancient world. It’s not that simulation hasn’t been used in radiology—strictly speaking, medical images are simulations of actual patients—but simulation-based training has valuable application in radiology.

To make the point that this age-old idea can be a valuable component of modern radiological practice and training, researchers at Louisiana State University wrote a book about the topic, Simulation in Radiology, which is scheduled to be published on June 1. Although the authors included copious research and modern learning theories in the book, they acknowledge that simulation-based medical training dates to antiquity.

“In medicine, you can find examples for simulation among the ancient Egyptians, Hippocrates, Galen, Avicenna, all actually doing some simulations related to healthcare to instruct their students or apprentices,” says coauthor Hugh J. Robertson, MD, a professor of clinical radiology at Louisiana State University Health Sciences Center (LSUHSC) in New Orleans. “So it’s certainly not something new. It doesn’t [exclusively] belong to this age. But if you look outside medicine, there is simulation in almost every aspect of life today.”

Not New but Necessary

Pilots routinely use flight simulators to learn how to handle specific situations and different types of aircraft. In the medical field, simulation-based training has been used to train surgeons in laparoscopy for more than a decade. The American Board of Surgery requires surgery residents to achieve Fundamental Laparoscopic Surgery certification, which is a simulation-based test, before taking the board examination. John Paige, MD, an assistant professor of clinical surgery and director of the applied surgical simulation department at LSUHSC and coauthor of the book, says simulation will become a mainstay of licensure for surgeons in much the same way that it is for airline pilots.

In recent years, researchers have developed a better understanding of how people acquire and master specific skills. That understanding helped make simulation an integrated component of clinical training. Researchers and educators are looking to apply knowledge and technology from areas such as anesthesia and surgery to radiology. The authors of Simulation in Radiology cite Swedish psychology researcher K. Anders Ericsson’s theory of deliberate practice to develop expertise as a significant influence on this thinking. The idea is to focus on critical tasks in a given field and re-create them in a way that allows the learner to practice and master them. Over time, more complex skills can be added until the learner attains the desired level of expertise.1

“We believe that when you [apply] the Ericsson model of delivering practice, [which is designed to] create experts, that you can do the same thing with mammography,” says coauthor Leonard Bok, MD, MBA, JD, chairman of the radiology department at LSUHSC. “We can probably create a much better mammographer, if we—in addition to the clinical training that we’re giving them in the mammography center every day—also add deliberate practice into that by giving them data sets [of mammograms with] more cancers than they would [normally] see because there are only two to five cancers in 1,000 mammograms in your daily practice. So by creating a deliberate practice step, where they can work on that on a day-to-day basis, we can create better mammographers than we would have under the classic circumstances.”

There are several factors that have increased the interest in simulation-based training in recent years. One is patient safety. Paige says the Institute of Medicine’s 1999 report “To Err Is Human: Building a Safer Health System” was a wakeup call for medical practitioners. A significant benefit of simulation-based training is that it provides learning opportunities without putting patients at risk.

Around the same time as the IOM report, the practice of minimally invasive surgery was becoming more prevalent. However, the training for minimally invasive procedures often took the form of a weekend course. Paige says increases in complications from these procedures highlighted a need for more rigorous, standardized training. Bok adds that the FDA gave a boost to simulation with its requirement that carotid artery stenting centers must provide simulation training, in addition to standard training, to earn accreditation.

More recently, changes to residents’ work schedules have reduced the number of hours they’re able to spend in clinic. Paige says many teaching programs opt to have their residents work on basic skills with simulators beforehand so they can maximize their time with actual patients. Simulation also helps residents develop familiarity and confidence in handling critical events that happen infrequently, which is a significant concern due to the reduced work hours.

Opportunities in IR

While simulation can be applied throughout radiology, it translates particularly well in interventional radiology (IR). Because IR is procedure based, simulation offers many opportunities to develop and refine clinical skills. Beginning with basic skills, such as tying knots, suturing, catheter placement, and the Seldinger technique, residents can then move on to virtual reality machines that simulate procedures such as carotid artery stent placement, aortoiliac and femoral arterial stent placement, or selective arterial catheterization. It also allows simulation of critical events that may arise, such as an embolism.

There are various types of simulators that can be used to teach these skills. They range from simple devices designed for a single skill, such as venous catheter placement or arterial puncture techniques, to complex virtual reality devices with computer interfaces that provide simulation of patient procedures while recording metrics to assess performance. Mannequins that display vital signs and respond to drug therapy are also useful, as are animal models and human cadavers. When the needed mastery has been achieved, the resident can begin to work on patients as part of a surgical team.

“There is a huge opportunity for simulation-based training in interventional radiology because that’s a procedure-based discipline, and also for emergencies in radiology, like CT contrast reactions,” says Paige. “It’s very useful for these low-frequency, very-high-risk events. A radiologist might not see it very often but, when they see it, they need to know how to react. The whole team needs to know how to react, and if the team can react together in a systematic way and they’ve practiced it—they’ve had that experience with the simulation prior to the event—the reward is great.”

At RSNA 2011, the authors outlined a potential protocol for simulation-based training. It begins with a didactic lecture that highlights the procedure’s hazards and potential risks that are being taught. The trainee must then pass a multiple choice test to demonstrate basic knowledge of the training topic. Then the trainee performs the simulated procedure with expert supervision and receives objective grading of his or her procedure skills. Next, the trainee takes a written posttest to assess his or her improvement in the subject knowledge and then participates in a private interview with the person who administered the test to receive advice on correcting errors. At this time, the trainee is asked to provide an opinion of the learning session. The training session is repeated over selected intervals and a logbook is kept to record performance over time.

In this model, the residents receive multiple opportunities over time to develop their skills until they can meet the established goals experts have developed. This provides time for reflection and increases the number of times that a trainee is exposed to the simulation, both of which are determinants of an individual’s ability to retain the information. However, the key to making the simulations effective, says Paige, is to develop proficiency-based goals.

“Just having simulation isn’t the answer,” says Paige. “It’s having a very-well-developed educational curriculum in which the simulator is used as a tool to obtain the learning objectives you have and having very-well-developed metrics or sets of tools that allow you to make sure that what you’re trying to teach is being learned.”

Looking for Validation

The development of simulation metrics is a work in progress. Although simulation is not new, its effect on radiology practice hasn’t been widely studied, and measuring the results of simulation-based training can be tricky. There are several variables that need to be considered.

“It’s a difficult science. Just to have a simulation device is a beginning. If it is used alone, it is only similar to a simple computer game without lasting and verifiable educational value. It has to be coordinated with advanced teaching and educational theories, psychology, physics, and engineering principles. There are a number of things that must be combined to make the use of simulation effective,” says Robertson. “So simulation is done as an algorithm, in which there are tests conducted before and after the simulation is done, and there are measurements of how the skills are [progressing]. It’s very complex. It’s taken years to develop, and it will continue to develop.”

Because medical practice relies on well-documented, verifiable data, some in the medical community believe that simulation-based training is a decade away from being a useful tool. The authors, however, are adamant that simulation is too valuable to set aside until then. They argue that there are numerous examples in other fields where simulation was effective even though it wasn’t scientifically validated. Bok points out that the Navy’s Top Gun program, which was instituted during the Vietnam War, was essentially validated by fewer pilots being shot down as a result of their simulation training. And Paige cites a satirical study that appeared in the December 18, 2003, issue of BMJ, which concluded that those who insist on rigorous evaluations for any and all medical interventions without regard to their observed benefits should participate in a randomized, controlled trial of parachutes.2

Although there haven’t been many randomized trials examining the efficacy of simulation-based training for radiology and IR, that doesn’t mean it can’t improve practice. The ACR Learning Center in Reston, Virginia, utilizes a training model that simulates a radiologist’s work environment. The course requires radiologists to look at hundreds of cases during a three- or four-day period on a 3D workstation.

Ultimately, the goal is to provide better patient care. Another potential advantage of simulations is that they may eventually allow specific patient data to be incorporated, marking another step in the quest for personalized medicine. Bok anticipates that, within the next few years, it will be possible to load a patient’s CT angiogram data into simulation software and practice an aortic stent placement procedure on a simulation of that particular patient’s aorta before it’s done on the patient. Robertson expects that simulation will include a patient’s diagnostic studies as part of pretreatment planning for other, more difficult procedures as well.

“Right now, we’re at the earlier stages of validation,” Bok says. “We put our residents through a program, and quite a few radiology programs do at this point, where they have eight or 10 different scenarios on a mannequin that might be a vasovagal response or it might be a major contrast response. When they go through that, they report that they learn from that, and they feel much more comfortable when they do have to handle a contrast reaction or vasovagal reaction in the department. What we ultimately want to show is that, by having them go through this process, reactions in the department are actually treated better and more quickly, and we have better patient outcomes. Once we are able to do that as a field on a large scale, then you’re going to see massive adoption of simulation techniques.”

— David Yeager is a freelance writer and editor based in Royersford, Pennsylvania, and a frequent contributor to Radiology Today.


  1. Ericsson KA, Krampe RT, Tesch-Romer C. The role of deliberate practice in the acquisition of expert performance. Psychol Rev. 1993;100(3):363-406.
  2. Smith GC, Pell JP. Parachute use to prevent death and major trauma related to gravitational challenge: systematic review of randomised controlled trials. BMJ. 2003;327(7429):1459-1461.