Video Game Technology Could Prevent Hospital Falls
COLUMBIA, Mo., March 19, 2014 — Video game technology may be able to serve a new purpose: preventing hospital patients from falling.
As many as 1 million people each year fall in U.S. hospitals, according to the Agency for Healthcare Research and Quality. And while falls can happen anywhere, patients in hospitals, particularly the elderly, are at increased risk of injury because of illness, injury or disorientation from medication.
Researchers at the University of Missouri-Columbia are studying ways to prevent these risks using video game motion-capture technology. They installed their system in each of six patient rooms at University Hospital. The device, which looks like a thin black box, easily connects to a computer and features sensors that detect movement and a depth camera that measures distance.
It collected data continuously, monitoring the rooms 24 hours a day, and sent a grid pattern of IR light into the room as a way of examining how objects and people distort the pattern.
"By seeing what happened before a fall, we can better understand what caused it," said Dr. Marilyn Rantz, who is also an RN, leader of the research team, and a professor at both the MU Sinclair School of Nursing and MU School of Medicine. "The more we know about what causes falls, the more effectively we can prevent them."
The device creates a 3-D map to show everything in a room and determine exactly what is happening — whether someone has fallen, or is simply picking something up off the floor, kneeling, etc.
Creation of a precise algorithm allows the device and computer to calculate the probability that the changes represent a person's fall.
Rantz noted that such technology can help health professionals learn about risk factors for falls and ultimately help create more effective ways of preventing them.
The researchers have explored other technologies for this purpose over the past several years, including video cameras, Doppler radar and sound sensors, said Dr. Marjorie Skubic, a professor of computer science at the MU College of Engineering.
Both Doppler radar and sound sensors can detect a person’s fall, but neither shows what happens leading up to the fall; and while video can see what happens beforehand, it can do so only when there is enough light.
The work was supported in part by grants from the National Science Foundation and the Agency for Healthcare Research and Quality. The research is published in the Journal of Gerontological Nursing
For more information, visit: www.missouri.edu