Vera Kettnaker Receives NSF CAREER Award: Research Aims to Keep Seniors Safe in Their Homes

Thomas Griffin

Most senior citizens prefer to live independently for as long as possible. The risks of in-home falls and injury, however, prevent many seniors from remaining self-sufficient. Vera Kettnaker, assistant professor of computer science at Rensselaer, has received a Faculty Early Career Development Award (CAREER) from the National Science Foundation (NSF) to develop a video monitoring system that may someday offer seniors a way to receive help automatically.

Kettnaker’s proposed “video-equipped intelligent environment” will be able to analyze an elderly person’s movement patterns to detect a potential problem and, if needed, summon help automatically. Her system could someday allow seniors to be safer in their own homes and self-reliant for many more years.

Kettnaker’s planned surveillance system would “learn” the pattern of a person’s regular activities during a two- or three-week training period. Using a mathematical model similar to those used for voice recognition and natural language processing, it would analyze the person’s locations and activities and how they change over time. The system would then be able to project expected or “normal” patterns of behavior for the resident individual.

The CAREER award provides a grant of $400,000 over five years and is the most prestigious honor the NSF presents to junior faculty. Kettnaker is one of 22 Rensselaer faculty members to receive this award in the past four years.

Current safety monitoring devices for seniors require injured or ill individuals to manually request assistance with the push of a button or a tug on a string. Such otherwise helpful devices are of no use following accidents such as falls or stroke that result in loss of consciousness.

Kettnaker’s planned surveillance system would “learn” the pattern of a person’s regular activities during a two- or three-week training period. Using a mathematical model similar to those used for voice recognition and natural language processing, it would analyze the person’s locations and activities and how they change over time. The system would then be able to project expected or “normal” patterns of behavior for the resident individual.

Kettnacker says it is much easier to track seniors than say, a teenager, since their life patterns are well-established. “Once you are 85-years-old, you’ve found your routine,” Kettnaker says. “You probably minimize trips around your home and generally have more structure in your life.”

For more, see press release.