Virtual Second Chance

Late at night, a hospital employee sits facing a bank of computer monitors. Flickering on the screens are vital signs beamed wirelessly from monitors at the bedsides of 64 patients—people whose lives may depend on his unwavering attention.

Then several things go wrong at once. Data flashing on-screen show one patient’s heart rate has plummeted, cuing a raucous alarm. As the operator tries to call a nurse, another patient flatlines—but is it cardiac arrest, or did she accidentally disconnect the monitor? Then an alarm goes off at another operator’s desk, and in the tumult, a fourth patient’s crisis goes unnoticed until it’s too late.

No need to fill out a death report, though.

Instead, Monifa Vaughn-Cooke (right), a UMD assistant professor of mechanical engineering, can change the specifications of this simulation with a few keystrokes—by giving the operator fewer patients to monitor, for instance. Or she can change the sound of the alarms to be less teeth-rattlingly distracting. With the system optimized to minimize stress and help operators perform at their best, next time around, maybe the patient will live.

With this “Matrix”-like ability to quickly switch the variables in real time—rather than simply guess about the effect of doing so—University of Maryland researchers can use virtual reality to study how the performance limits of telemetry operators affect patient safety.

“In the real world, the ability to go into a safety-critical environment and change design parameters is limited,” Vaughn-Cooke says. “Distracting operators while they’re performing their jobs would be potentially hazardous to patients’ lives.”

But here, she’s free to experiment.

Nearby, two of her students are prepping the virtual reality “cave” in the Hybrid-System Simulation and Integration Lab, which she co-directs. When they turn it on, a slightly cartoonish version of the telemetry monitoring room at Medstar Washington Hospital Center flashes onto the translucent walls of the small room lit by hidden video projectors. Virtual co-workers stroll through, patient data light up computer screens, alerts blare.

The operators—student stand-ins for now, but actual Medstar hospital employees later this spring—don 3-D glasses, a glove with motion sensors that allow them to interact with the virtual world, an EEG headset and a chest band that monitors their own vital signs.

With this setup, Vaughn-Cooke can ramp up the pressure, adding more demands and requiring higher and higher levels of performance from operators. She doesn’t have to ask the operators how stressed they are, because the neurophysiological sensors objectively track how they’re reacting in real time.

“Eventually you place a burden on them that exceeds their cognitive ability,” she says. “They’re in a situation where they have more than they can process at one time, and this is when alarms are unintentionally ignored or there’s a delayed response.”

The performance and physiological data she gathers should give insights into how workstations and technology can be optimized to improve performance. Among other possibilities, if an operator’s brain activity and vital signs indicate overload, the monitoring system might be programmed to automatically send some of a harried worker’s responsibilities to another facing fewer responsibilities.

Overloaded telemetry operators can be a genuine hazard to hospital patients. According to a 2010 Boston Globe investigation, over 200 people around the country died over the previous five years because of failures in telemetry monitoring.

The technicians who monitor patient data take their jobs seriously and do them well, says Royce Payton, who supervises nearly 40 telemetry technicians at Medstar Washington Hospital Center. But they can’t control the stressful circumstances they face.

“It can always get a little crazy, a little hectic,” he says. “Sometimes we might have two or three [emergency] codes at one time, or multiple patients off the monitor.”

He’s glad about the hospital’s collaboration with Vaughn-Cooke, and hopeful the research can find ways to ease sometimes-heavy demands on his workers.

For instance, he’s interested in whether the virtual reality testing can demonstrate that streamlining technicians’ paperwork demands might result in more effective patient monitoring.

“I’d like to see the effect of the simulator taking some of that writing off of us,” he says.

Vaughn-Cooke’s students are fascinated by the varied subject matter that virtual reality research thrusts upon them. Electrical engineering major Christopher Chang ’15 says an interest in video games initially drew him to research in uses of virtual reality. In Vaughn-Cooke’s lab, he’s had the opportunity to put the technology to use for serious purposes—not only improving patient telemetry, but also in another project using virtual reality in a smoking cessation program for people with mental illness.

“These are really interesting, hands-on things you can’t find in a classroom,” he says.

Vaughn-Cooke’s performance testing coupled with neurophysiological monitoring has wider applications, and could be used in training power plant operators, air traffic controllers and others with high-risk jobs.

“Usually the operators sit and stare at a screen without any alarms for a stretch of time,” she says. “But when things do begin to happen, pretty often they pile up one on top of the other. This is when patient safety is compromised.”