"Virtual OR," Feature Article, November 2006

virtual OR

Engineers are developing systems that enable future healers to practice surgery and other skills on model patients in simulated spaces.

by Alan S. Brown, Associate Editor

Cool heads prevail in emergencies, and the coolest heads rest on experienced shoulders. It's the "been there, done that" attitude that comes from time, trial, and error.

But when the stakes are life and death, as they often are in operating and emergency rooms, no one is going to let medical students or residents learn from their mistakes. That's why the medical profession is turning to engineers to develop tools that give doctors crash courses in facing life and death decisions as part of an operating team.

"There's a coordinated sequence of events that has to take place in an operating room," said Leonard Weireter Jr., a surgeon at Eastern Virginia Medical School who also heads Sentara Norfolk General Hospital's Shock Trauma Center. "If the anesthesiologist, surgeon, and circulating nurse know how to communicate with one another, a lot can get done." If not, the ballet devolves into something more like a mosh pit. It can happen quickly in an emergency, such as cardiac arrest or allergic reaction. It may also occur if a member of the operating team misreads an instrument or misunderstands a command.

The virtual operating room allows real surgical students to interact with virtual instruments, and work with virtual surgeons, nurses, and anesthesiologists.

Today, medical students and residents learn to cope with life-threatening emergencies by living through them. They stand beside skilled practitioners, the same way apprentices learned from their masters for hundreds of years. They watch, assist, and ultimately take the scalpel into their own hands under the watchful eye of an experienced surgeon.

Yet even the busiest hospital presents only a limited range of experiences for any given type of operation. While human beings vary widely, most operations follow a routine set of procedures. Doctors may go through years of medical school and residency, and never encounter more than a handful of true emergencies. They may never confront a life-and-death decision until they are out on their own.

That may be about to change.

Taking a lesson from other professionals who must sometimes face critical decisions—pilots, chemical and nuclear power plant operators, or military officers—the medical profession has begun to use advanced simulations and mechanical feedback to train doctors. The work is still in its very early stages. Yet the new technologies promise interactions that will blur the distinctions between reality and models in virtual space.

Imitation of Life

The concept is simple. Doctors, nurses, and paramedics learn and practice procedures on simulators until they become proficient. The simulators then vary symptoms to depict medical emergencies that most medical personnel rarely encounter. Future doctors, for example, can rehearse emergency procedures the same way pilots use simulators to learn how to pull out of a spin or fly with a damaged engine.

Over the past decade, several companies have introduced simulators. Most consist of a plastic and rubber mannequin, a haptic feedback system that simulates the resistance of medical instruments moving through the body, and imaging systems that show the locations of the instruments.

Today's medical simulators have limitations, but they are moving rapidly into the medical mainstream. Two years ago, for example, the Food and Drug Administration approved a carotid stent, developed by Guidant Corp. of Indianapolis, that expands blocked arteries in the neck. Before doctors could perform the risky procedure implant, the FDA required doctors to undergo four hours of simulator training.

"This is the first time that FDA required simulator training," said Mark Scerbo, a professor of psychology at Old Dominion University in Norfolk, Va., and co-director of the National Center for Collaboration in Medical Modeling and Simulation. "This may be the start of a new model for training doctors. Simulations can also let us test new devices and procedures without putting patients at risk."

Scerbo is at the forefront of those changes. As a human factors psychologist, he has studied how doctors learn their craft. He is quick to point out the flaws in existing simulators. Each system covers only one specific type of procedure, such as gall bladder removal or ectopic pregnancy (where a fetus grows outside the uterus). While some simulations are realistic, others are not. All are expensive and usually carry six-figure price tags.

More significantly, today's simulators reproduce only a handful of emergency conditions. None teaches the critical thinking and teamwork skills needed inside an operating room.

Scerbo's National Center for Collaboration in Medical Modeling and Simulation wants to change that. It was formed four years ago when Old Dominion's Virginia Modeling and Simulation Center, which has close ties with the military, teamed with neighboring Eastern Virginia Medical School.

On one hand, the center evaluates existing simulations. "The biggest question medical schools have before they invest a few hundred thousand dollars in a simulator is, ÔDoes it work?' There's no empirical evidence that one is better than another or whether any of them are effective," Scerbo explained. His goal is to quantify their efficacy.

The center also hopes to commercialize new technologies. Its debridement system, for example, uses virtual reality to walk students through cleansing large surface wounds. "A person can come in, practice a skill on a simulated limb, and receive feedback from the system," Scerbo said. "The first time that person sees a patient, he or she can perform the procedure."

Finally, the center is building a comprehensive operating room simulator. The researchers have built a complete operating room around two existing procedures, gall bladder removal and an ectopic pregnancy. Inside that virtual environment, medical students can interact with simulated doctors and nurses while they operate on a mannequin. The unit is intended to train doctors in both critical thinking and teamwork.

The modeling environments and haptic feedback devices now being adapted for surgical training have existed for decades. Why are physicians only just beginning to tap their power? The answer, Scerbo said, involves litigation and changes in operating room practice.

"Medicine is a lightning rod for litigation, and anesthesiology is one of its riskiest specialties," he said. Starting in the late 1980s, anesthesiologists teamed with engineers to reduce operating room errors. They developed training mannequins that simulated such physiological responses as high blood pressure and choking.

A debridement center uses haptic feedback to provide medics with a realistic experience as they remove debris and cleanse a virtual wound.

"One of the most serious issues in the field is intubation, getting a tube down the throat without choking the patient," Scerbo said. "The more you do it, the better you learn. Doctors joke that the reason they call it a practice is because they practice on you and me. With mannequins, they're learning on a device and not on a patient."

The use of mannequin simulators coincided with the advent of minimally invasive, or laparoscopic, surgery. Instead of slicing open a body, surgeons inserted cameras and surgical instruments attached to long rods through small incisions. They then performed the procedure guided by camera displays of the organs.

Minimally invasive surgery reduced recovery times dramatically, but proved difficult to learn. "It's like doing very sophisticated surgery with chopsticks in your hands," Scerbo said. "It takes a lot of training to look at a two-dimensional display and understand what your instruments are doing. There's a real need to train doctors, and not on patients."

Working laparoscopic instruments takes more than looking at a video monitor. It also requires a sense of touch. Off-the-shelf haptic feedback devices reproduce the forces laparoscopic instruments encounter in the body.

Haptic devices provide force feedback. In surgical simulations, they are typically robotic arms that work in reverse: Instead of applying force to an object, they provide force feedback when someone moves an object. When a student moves a clamp at the end of a robotic arm, the haptic system calculates the amount of force to apply against that motion by gauging how the scalpel interacts with a computer-generated model of tissue in which it moves.

"It's very difficult for a haptic device to replicate what the skin senses, such as the sensation of picking up a tennis ball in your hands," Scerbo said. "It's much easier to replicate the resistance of a rod moving through a body."

Faithful Enough

"For years, we thought medical simulator haptics had to be incredibly precise," Weireter said. "We talked to the Air Force about their high-fidelity models of airflow over an F-16, but the shape of a liver is far more complex than a wing. But we found we didn't need to spend a billion dollars to create high-fidelity haptics."

In fact, students typically learn laparoscopic surgery using low-tech devices. They simply poke their camera and instruments through holes in a black box and practice hand-eye coordination skills, such as transferring objects from one hand to another and tying knots while watching a video display. Simulators eventually add haptic feedback. "It turns out you don't need the high-fidelity haptics," Weireter said. "It's the repetitive practice of the motion that counts."

Yet haptics plays a large role in the center's debridement system. Debridement is the system for cleansing wounds that are too large to stitch closed. Medics, paramedics, and nurses must learn to clean the wound and remove dead tissue, glass, and other foreign objects to prevent gangrene and infection.

"It's really a simple procedure," said Hector Garcia, a Virginia Modeling Analysis & Simulation Center research scientist. "If you can use a fork and knife to cut chicken, you can do this. But we don't want to have to take a medical doctor's time away from other tasks to teach this simple procedure."

The debridement system attempts to replace a physician with instructional materials and simulations. First, a virtual instructor describes different types of wounds and lacerations. Then it shows videos of procedures. Finally, the system walks the student through the cleansing of a wound containing glass shards or other objects by using a three-dimensional simulation projected onto a large reflective screen.

No one would mistake the virtual wound for the real thing, but it has enough fidelity to give students practice. "They grab the grasper-type tool affixed to the end of the robotic arm and use it to clean the wound," Garcia said. "The robot has six degrees of freedom and the ability to provide resistance or deny movement in any direction."

As the robotic arm moves, it interacts with a computer-generated wound. The computer represents the skin's surface as a mass-spring model, a mesh of nodes connected by lines. Each node has a mass associated with it. The lines between them act like springs. When the instrument touches a spring, the model calculates the resistance based on the mass of the node and the resilience of the spring. This calculation determines the haptic resistance of the robotic arm.

"Some surfaces deform and bounce back when pushed," Garcia said. "Others offer more resistance. It doesn't behave like real tissue, but our model is based on a more precise and computer-intensive model of how skin deforms. It's close enough to give the illusion of skin, but simple enough to run in real time on our computers."

According to Weireter, "It's a great device, intended to teach real novices how to clean up a sophisticated wound so you can move the patient safely." Weireter and other team members are now looking at ways to make the
debridement simulator generate a broader variety of wounds and teach students to monitor them for signs of infection after treatment.

Virtual Operations

The virtual operating room creates even more complex interactions between real and virtual space than the debridement system. The space itself is a combination of the real and the virtual. High-intensity lights glare down on a mannequin lying on the operating table. The room's walls display virtual monitors, instruments, and a transfusion kit. Two live students share the room with simulations of other medical professionals.

"The room combines psychology and engineering," Scerbo said. "If you look at advances in safety made in other high-risk domains—aviation, nuclear power, military operations—they were achieved by people who understood the entire environment in which they perform. They understood their tasks, their tools, and the role of their coworkers.

"Doctors and surgeons don't perform individually. They perform with other doctors and nurses, with instruments and displays, and often with lack of sleep. They may go in for a 90-minute procedure, but wind up standing through a four- or five-hour operation."

The new technology promises interactions that will blur the distinctions
between reality and models in virtual space.

Decision-making and communications are critical in that environment. "First-year surgeons learn procedures, but then they have to understand the interaction of drugs, operating room conditions, and patient status," Scerbo said. "If something unexpected happens, they have to be able to handle that, too."

Today's surgery simulations teach only procedures, he noted. None shows doctors the context in which they have to perform. Scerbo's goal is to take existing skills-oriented simulations and then add operating room interactivity.

"What we've done," added Weireter, "is put a simulation that teaches technical skills into an interactive environment, where the other people in the room are not real people but virtual images. We're not going to teach you to do the operation, but how to act with other people so you know how to interact when catastrophes occur. Instead of making it up at the line of scrimmage, we're going to drill team behavior so that when something happens, it's no big deal because we've prepared for it."

As the virtual operating room evolves—and this may take years—it is expected to drill students and residents in critical thinking and communications skills. They will see more and more varied emergencies than the cases that come through the hospital doors when they are on shift. They can also schedule virtual team practice at their own convenience.

Into the World

Equally important, the virtual operating room gives human factors researchers a tool to study how and why surgeons make mistakes. "We can look at the sources of errors that creep into procedures and design countermeasures," Scerbo said. "We can build better systems that match the capabilities of human users without overloading or underloading them."

Scerbo, Weireter, and Garcia freely admit that virtual surgery is still in its infancy. Many commercial systems have design flaws or leave out critical steps. A simulator that's designed to teach how to draw blood, for example, doesn't let doctors or nurses feel an arm to get a sense of a vein's location. "People who trained on that system did worse when they went to take blood than those that trained conventionally," Scerbo said. "It was like learning to fly on a flight simulator that doesn't let you fly in the wind."

Yet simulation systems have already scored victories. Several years ago, members of the U.S. military's Special Forces challenged Weireter to use his system to solve a battlefield problem. "The medics we trained performed great in well-lit rooms with elevated operating tables," he recalled. "But they didn't know how to perform when people were shooting at them in the dark.

"So we took medics and put them in an environment where they had to keep their heads down or they were shot by a sniper. When they mastered that, we turned off the lights so the only light they had came from explosions. We showed we could train them to perform in that environment, to focus on what's important, and keep their heads down so they didn't get shot."

Those medics are now saving lives in Iraq. They are not succeeding because their medical skills are different from the medics who trained before them. Instead, they save lives because they understand the context in which they must put their skills to use.

One day, thanks to medical and surgical simulations, that might be true of all doctors and nurses.