This Sensor System Predicts Your Risk of Falling Down

The new in-home sensor system measures gait speed and stride length to identify adults who are at high risk for falling, and alert healthcare providers to intervene before a fall occurs.

Kristopher Sturgis

University of Missouri researchers have designed a sensor system that can predict likely falls.

The imaging system, installed around the living space, to not only detect when falls occur, but recognize the signs of risk to prevent falls from ever happening, says Marjorie Skubic, professor of electrical and computer engineering at the university and director of the Missouri Center for Eldercare and Rehabilitation Technology.

"We have developed fall detection systems before, but a real win is when you can detect that someone has a high risk of falling, and get them help that prevents the fall in the first place," Skubic says. "A fall, or even the risk of falling, is often the turning point that forces older adults out of their home. It can have a huge impact on their independence and quality of life. If someone falls and cannot get help soon, this also impacts their chances of recovery. We have seen this with our own family members, so we want to detect falls and get people help as soon as possible."

The group worked with sensor systems in place at TigerPlace, an aging-in-place retirement residence in Columbia, MO, that works with the university to aid in the development of innovative senior care technologies. The system works by generating images and an alert email for nurses and caregivers, indicating when irregular motion is detected. This information can be used to assist nurses in assessing functional decline, and enable intervention that can provide treatment to prevent falls that result in serious injury.

As Skubic and her colleagues move forward with the sensing technology, she says that the main goal is to continue to refine a system that can predict and identify risk before it happens.

"The in-home gait analysis system is part of a larger system that recognizes very early changes in health and generates alerts," she says. "This is designed to get people help very early, to address health problems when they are still small and manageable."

Early results from an analysis of the sensor system found that a speed decline of five cm per second could be linked with an 86.3% probability of the patient experiencing a fall within the next three weeks. Additionally, the team found that shortened stride length could be linked with a 50.6% probability of falling within a three week period.

Ultimately, the research indicated that implementing the new sensor technology at TigerPlace enabled residents to live independently for four years on average, compared to the national average of just 22 months.

Despite the success of the in-home system, Skubic says that there are no plans to develop any kind of wearable technology for older adults, as the group aims to keep the system as simple as possible as they begin to roll the it out for commercial use. 

"Our research has shown that many older adults do not want to be bothered with wearing sensors," she says. "They are often unable to manage the charging of wearables and use them consistently. By embedding sensors into the environment, they do not have to do anything special. We've made the technology available now, and Foresite Healthcare has licensed the technology and has begun installing it in senior housing."

Kristopher Sturgis is a contributor to Qmed.

Like what you're reading? Subscribe to our daily e-newsletter.

[Image courtesy of University of Missouri]