Play Live Radio
Next Up:
0:00
0:00
0:00 0:00
Available On Air Stations

Real Life 'Marauder’s Map' May Help Improve Patient Care in Nursing Homes

Researchers at Carnegie Mellon University have developed a way to track the locations of individuals in complex, indoor settings such as nursing homes.

Developers liken it to the Marauder’s Map featured in the Harry Potter books and movies, which allows Harry Potter to see anyone’s location at Hogwarts School of Witchcraft and Wizardry.

But instead of magic, this system uses a network of cameras and algorithms to track movement. Researchers said this could be important in keeping track of residents of nursing homes.

“Especially since staffing levels tend to be very limited, can we maybe help caregivers to alert them to things that are maybe declining earlier than they would notice on their own,” said CMU researcher Alexander Hauptman.

The system was tested tracking 13 residents at a facility. It can identify if a patient sits in front of TV a little longer than normal each day, something each nursing shift would not necessarily notice, but that could  over time indicate a trend that the staff needs to know about.

“Or we’re finding they’re more agitated, or they have fewer social interactions than they did previously — those are signs of depression that would be good to know ahead of time, as early as possible, so they could deal with them," Hauptman said. "Potentially, some of those things are side effects of medications that could then be adjusted appropriately.”

In light of the ongoing National Security Agency scandal, however, researchers realize this sort of technology may not be welcomed in many settings.

“It’s a very difficult thing because you really have to balance whether anything positive that comes out of it is worth the amount of privacy you’re losing,” Hauptman said. “We are currently, actively working on methods to make people unidentifiable, but we still recognize where they are and what they’re doing.”

That may be a challenge, as facial recognition is a key component in the tracking. The findings will be presented at the Computer Vision and Pattern Recognition Conference in Portland, Oregon later this month.