Smart Tools for the Elderly
“Your mother is losing her memory,” the doctor said. At 80 she was fit for her age. But a series of “mini-strokes” had impaired the area in her brain that was home to short-term memory. She would become progressively more disoriented. She might forget to take her medications and to fix her mealsor to shut of the gas burners on the stove. It was time, my family agreed, for her to move to a retirement home. But she hated the idea.
My mother’s story is all too familiar to millions of older Americans, for whom the years we call “golden” are tarnished by illness and senility. Three quarters of Americans over 65 suffer from a chronic disease. And in the U.S. and other industrialized societies, many adult children are too busy making a living and raising their own families to care for their elderly parents at home.
Dr. Robin Felder, director of the Medical Automation Research Center at the University of Virginia, has given a lot of thought to that problem. He and his Healthsense Project team work to extend senior citizens’ years of independence with affordable high-tech assistance devices that help older users to keep an eye on their environment-and their environment to keep tabs on them.
The Seeing Eye Walker
Senior citizens comprise three out of every four blind Americans. To help this population retain their freedom of movement, Felder’s team developed a tool they call the Smart Walker (See “Smart Walker Strolls Ahead,” technologyreview.com July 5, 2001). The walker is a three-wheeler fitted with a laser scanner. Onboard software processes the data from 180 degrees of approaching terrain and steers the front wheel toward openings and away from obstacles. The walker, known as a “passive shared-control navigation system,” does not lead users around against their will. Instead, mobility is a result of give-and-take between the user’s self-propulsion and the walker’s automated reactions. If, for example, a user does not see a coffee table ahead, the walker will detect it, override the user’s steering to avoid it, and thereby prevent a possible fall.
The Healthsense Project is developing a second tool, to complement the walker and help seniors to continue to live in their own homes-known to gerontologists as “aging in place.” The goal is to build a network of sensors in a user’s home smart enough to recognize ordinary patterns-such as eating, sleeping, and greeting visitors-and to alert caretakers to out-of-the-ordinary ones-such as prolonged inactivity or absence. “For instance,” says Felder, “if you go into the bathroom then disappear off the sensor for 13 minutes and don’t show up anywhere else in the house, we infer that you took a bath or a shower.” But, he continues, if a person falls and remains motionless for 20 minutes, “the system would record that seismic event and instantly somebody would be notified.”
Felder’s team has installed a prototype in the Charlottesville, VA home of Jim Humphries, a registered nurse and friend of Felder. Though only 43, Humphries volunteered to be the test subject in April 2001. Redundant detectors record Humphries’s movements in different rooms: motion detectors monitor his lateral movements while infrared ones record his movement directly toward the sensor.
Sensors abound in the kitchen and bathroom, where they’re attached to cabinets, drawers, appliances, and floors. “The kitchen has sensors that are triggered when I open the microwave or the refrigerator,” Humphries explains. “There are pads [with sensors] in front of the stove andsink, so they can pretty much infer when I’m cooking a meal.” Likewise, a sensor pad beside the washing machine tells the system when Humphries does his laundry.
While the sensors are the most obvious element of the In-House Monitoring System, they are useless without software to interpret their data. A computer in Humphries’ study, explains Felder, “synthesizes the data from all the sensors and then sends it off through the Internet to the neural network-based central monitoring system.” The central monitoring system-a host computer in Felder’s University of Virginia lab-then refines the data and sends reports to a doctor or caregiver.
The Virtual Physical
Felder is working to improve both systems. He plans to add “drop-off detection” to the walker so it can see height changes in walking surfaces like stairs. He has tested sensors that measure how a subject walks. One day, such “gait monitors” might note that their user has developed a slight limp and report that information to the caregiver before the condition gets serious. Felder is also experimenting with a bed pad that measures a sleepers’ blood pressure, quality of sleep and breathing. In the longer term, Felder foresees ways to track what goes into users, as well as what comes out. “Every time you’d use a package you’d scan it,” he says, “and we’d make it so convenient that it would be like those readouts at the grocery store.” In the bathroom, he imagines users could just put a device in the toilet to automate urinalysis.
While both devices can be used by anyone, they were developed especially for lower income seniors. Felder estimates that the Smart Walker and Smart In-Home Monitoring System will each cost between $300 and $400, and will be covered by health insurance.
But don’t all these sensors come with a hidden cost-privacy? Humphries, the first man to live within the system, says no. He insists that the system, with no cameras or microphones, is “as minimally invasive as possible.” And he is adamant that the benefits far outweigh the cost. “My father died of a heart attack,” he explains. “He was home alone unmoving for two days before a friend found him. A system like this could have saved his life.”
The inside story of how ChatGPT was built from the people who made it
Exclusive conversations that take us behind the scenes of a cultural phenomenon.
Sam Altman invested $180 million into a company trying to delay death
Can anti-aging breakthroughs add 10 healthy years to the human life span? The CEO of OpenAI is paying to find out.
ChatGPT is about to revolutionize the economy. We need to decide what that looks like.
New large language models will transform many jobs. Whether they will lead to widespread prosperity or not is up to us.
GPT-4 is bigger and better than ChatGPT—but OpenAI won’t say why
We got a first look at the much-anticipated big new language model from OpenAI. But this time how it works is even more deeply under wraps.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.