From screen … to spleen: med students and experienced surgeons alike benefit from a 'flight simulator' for doctors

Insight on the News, Nov 4, 1996 by Phil Berardelli

Med students and experienced surgeons alike benefit from a `flight simulator' for doctors.

The patient is suffering from a clogged iliac artery in his pelvis, cutting off blood flow to the lower extremities. Surgeons will correct the problem using laparoscopy, a technique that allows them to operate with tiny sheaths and catheters, avoiding invasive cutting. Insurance companies prefer it because it reduces costs and recovery time; patients like it because post-operative pain is lessened. But the procedure has one drawback-it is difficult to teach to medical students.

Laparoscopy cannot be performed on cadavers, and experience gained by practicing on animals has limited value. Up until now, surgical techniques have evolved through trial and sometimes costly error (surveys by medical journals show that surgeons need up to two dozen patients before they master a new procedure). And since complications are unpredictable as well as undesirable, student experience with them usually is sporadic.

But a new computer system, the equivalent of a flight simulator for doctors, aims to solve these problems. Unlike flight simulators, however, "the terrain of the body, the internal organs, must allow for interaction," says Gregory L. Merril, president of HT Medical Inc. of Rockville, Md., which developed the system. "The organs must be programmed with behaviors and basic principles of physics so they respond appropriately when they are cut, tugged and stretched."

Merril, a psychologist and biologist who formed the company in 1987, says the new teaching tool was made possible by the onset of "reality engines," workstations manufactured by Silicon Graphics Inc. that can process data 5,000 times faster than personal computers. (The same machines helped create special effects for Jurassic Park. ) The computers run Teleos, software developed by HT Medical based on a massive photographic database derived from detailed CAT scans (see sidebar).

The fact that laparoscopy uses video -- surgeons concentrate on the monitor, not the patient -- makes the simulations all the more real. At the beginning of each operation, students approach an artificial human torso lying on an operating table. They insert instruments into a preselected incision. (Sometimes sheathed instruments are attached to a metal frame and manipulated like joysticks.) Once inside the body, however, when a student must rely on the monitor, the operation becomes an exercise in virtual reality. The simulated tissues respond to the slightest movement of the instruments: Sharp edges cut them, blunt surfaces push them aside, needles puncture and sutures bind. Students even get to react to computer-controlled tactile feedback.

How real is it? "So real it's frightening," says John Kaufman, an assistant professor at Harvard University's school of medicine, who participated in the system's development. Virtually real. "The only difference is that there is a computer in the middle."

Kaufman's Harvard colleague, Steven Dawson, also believes the system has great value. "Right now, if a complication occurs, the only people who can learn from it are the ones in the operating room." But by creating realistic simulations, "we can teach many trainees what happens at a given point and how we get out of that situation."

According to Dawson, the system also will benefit patients who face difficult procedures. In such cases, a patient's X-rays or scans could be added to the simulator's database. "The medical team can practice over and over again until they're prepared to handle any complication," he says.

RELATED ARTICLE: 3-D Anatomy

Science has been looking to create a complete anatomical database of the human body to update the diverse collections of photographs and drawings that have been used for decades.

The effort nearly is completed. It took five years, most of which time was devoted to finding the "perfect" cadavers for the extremely detailed dissection and image processing.

According to officials at the National Library of Medicine, or NLM, which sponsored the Visible Human Project, the specimens had to be 20 to 60 years old, less than 6 feet tall, of normal weight and have "no traumas, significant surgeries or visible abnormalities." They could not have died from a disfiguring accident or illness. And, of course, the individuals had to be willing to donate their bodies to science. The cadavers finally chosen were those of a 39-year-old convicted murderer who had been executed by lethal injection and a 59-year-old woman who had died suddenly of heart problems.

The work was conducted at the University of Colorado's Health Sciences Center. Both cadavers were given complete MRI and CAT scans within 12 hours of death. The bodies were frozen in polymer foam and cut into four quarters, which were embedded in gelatin to stabilize their structure. As digital cameras recorded the process, a machine called a cryomacrotome planed extremely thin sections from the bodies. When the process was completed, the remains were cremated.