Microsoft Kinect cameras are set up in the apartments of volunteers in a residential care facility in Columbia, and researchers are studying how the systems can prevent falls, serious illness and accidents.

It’s the same technology used in X-Box video gaming systems, but the residents aren’t exactly using them to play Black Ops.

Instead, the Kinects are helping gather data on how to keep the elderly safer in their homes, and stay in their own homes longer.

Computer engineering professor Marjorie Skubic is working with students to create algorithms to evaluate what the Microsoft Kinect cameras detect.

She stresses that this is different than taking still photos or videos of the residents at Tiger Place, which is important to note since senior citizens can be leery of privacy invasion.

“We do not have anyone looking at camera images,” Skubic says. “We’re not even saving camera images.”

Rather, the Kinects produce an infrared depth image that’s best described as a three dimensional silhouette.

Skubic is working with doctoral student Erik Stone to study how the Kinect can monitor behavior and routine changes in residents, changes that can indicate increased risk for falls or early symptoms of illness.

In addition to falls, the systems can monitor a resident’s gait. As Skubic explains, when someone’s gait changes, it can indicate an illness or a cognitive disorder.

She gives the example that urinary tract infections, which are common in older adults, can present in unusual ways, and even appear to be dementia. If the system can be programmed to search for such signs, a doctor can then easily test for a UTI, prescribe antibiotics and put the patient on the quick road to recovery.

Conversely, if a UTI continues to go undetected, it can lead to bigger and more serious problems.

Another doctoral student, Liang Liu, is collaborating with Mihail Popescu, assistant professor in the Department of Health Management and Informatics, to develop a fall detection system that uses Doppler radar to recognize changes in walking, bending and other movements that may indicate a heightened risk for falls. Different human body parts create unique images, or “signatures,” on Doppler radar. Since falls combine a series of body part motions, the radar system can recognize a fall based on its distinct “signature.”

“Falls are especially dangerous for older adults and if they don’t get help immediately, the chances of serious injury or death are increased,” Liu says. “If emergency personnel are informed about a fall right away, it can significantly improve the outcome for the injured patient.”

The sensing systems at TigerPlace provide automated data that alert care providers when residents need assistance or a medical intervention.

Skubic says the system allows residents to maintain their independence and take comfort in knowing that illnesses or falls may be detected early. The ultimate goal, she says, is to make this technology available on the open market so that older adults can live at home longer and have the security of knowing if something happens, help is on the way.