Personal health and wellness technologies are projected to be a $5 billion business this year, according to the Consumer Electronics Association.
Even President Barack Obama wears a wearable wellness device — a Fitbit — on occasion.
But, as it turns out, wearable technologies have a big obstacle to overcome: sensors — the miniaturized devices that measure things like speed and motion.
They are technically called MEMS, or microelectromechanical systems. We are most familiar with them in the forms of the accelerometers and gyroscopes that help smartphones and tablets keep track of motion. This is how we play games on our smartphones by just tilting and moving them, and how wearable wrist trackers count up our steps.
“The sensor market today is being driven by mobile technology,” Charlene Marini, vice president of marketing of embedded segments at ARM, said earlier this year, in an interview with Marketplace at the Consumer Electronics Show. “There’s a huge volume, of course, in mobile. And so mobile technology is being reused in things like wearables.”
That reuse has had its limits. In wearable devices, current MEMS sensors have not been able to keep up with some of the rigors imposed on them. For example, they have not been as power efficient as needed.
They also do not perform all the functions that wearable device designers dream up.
“A lot of these wearables end up in a sock drawer,” says Karen Lightman, executive director of the MEMS Industry Group. “After you watch your steps for one week you’re like, ‘Yeah, I get it. I understand. I need to walk more. OK, that’s not helpful.'”
But what if you could track more than just your steps or your heart rate? There are efforts to do that by addressing the shortcomings of current MEMS sensors in key ways: building new sensors, creating software that better operates those sensors in more rigorous conditions, and better understanding the data coming out of current sensors.
The Chicago-based startup Rithmio is taking the latter approach. You can see the function of a program they’ve created in the video below:
Rithmio co-founder Adam Tilton is working on a program that can track all exercises by learning the unique patterns of movement from each exercise, as performed by each individual user. In a demonstration, he showed how Rithmio’s program could recognize a new weight-training exercise within four repetitions, consistently, in multiple attempts.
“So, if you give me 10 seconds of motion-sensing data, I’ll tell you whether the user did 10 jumping-jacks, or five bicep curls or whatever,” Tilton says.
Tilton could make his program even more accurate, if he had more sensors — ones embedded in clothes. But put those clothes in a washing machine, and the sensors would be ruined. Current sensors can’t be washed.
“There’s still issues with respect to interoperability. There’s still issues with respect to energy and … power management,” says Lightman. “And for wearables that’s a big deal.” That’s because consumers will want wearable devices (even smart shoes and T-shirts) that don’t need recharging too often.
There’s a race to overcome these limits. Recently, Samsung came out with a new all-in-one chip that’s more energy-efficient and has better communication. Intel has announced a similar chip.
Merini says sensors for wearables are quickly evolving, and within a few years there will be a “greater use of new types of sensors, that might not be applicable to phones. For instance, body type sensors, heart rate sensors, etc.”
Mehran Mehregany, an engineering professor at Case Western Reserve University, is keeping track of technological advancements in wearable sensors. He says there are efforts to improve sensor technology so that they can measure not just our external activities and surroundings, but also what’s happening inside our bodies.
“For example, it would be fantastic to do long-term, non-intrusive blood glucose monitoring,” Mehregany says. “The sensor technology is not there to be able to do that type of measurement reliably.”
Mehregany also points to accurate, reliable blood pressure monitoring, which he says is the “Holy Grail” for wearable sensors. Right now, MEMS sensors for consumer-grade wearable gadgets still can’t reliably perform that measurement either, Mehregany says.
But, he predicts that in 25 years, sensors will measure many vital signs and will be embedded in us, acting like a biological black box. Just as black boxes today keep track of the mechanics of an airplane.