hello junko.yoshida. 3 years ago I got a fitbit to track activity whilst recovering from a knee injury. I then got the Striiv Play because it makes walking part of a fun game. Over time, observation of the data and the use of APIs to sync the data to various website accounts such as runkeeper. fleetly, hubub, achievemint, myfitnesspal, everymove, fitocracy, loseit, mapmyfitness, walkertracker, and so on, became addictive! My kids started using devices that I didn't have so I got more devices to compete with the kids. Some of the activity tracking websites incent one by offering prizes and cash, which is how I got the Bodymedia Fit LINK (awful device), the Jawbone UP24 (awful device) and the Omron USB. You can see the data disparity best via my everymove account. Browse to https://everymove.org/ and look for my account name "Michael Baker" to see the huge disparity between the devices.
When I started working out I looked for a device to aid me in quantifying my development and looked through various studies and articles to find the best one. I'm just curious which studies you looked at as the Bodymedia Fit I bought from an ISU study is within 1% accuracy, good enough for me. I learned of it from the IEEE Spectrum Article "How I Quantified Myself" August 30, 2012:
Gregory Welk, director of the Nutrition and Wellness Research Center at Iowa State University, has tested the accuracy of this new generation of wellness devices. In 2011, Welk strapped eight activity trackers and one lab-grade monitor onto each of 51 volunteers. For 71 minutes the volunteers ran on treadmills, swept floors, biked, and walked outside with heavy backpacks while the gizmos measured their energy expenditure. The most accurate commercial unit, Welk found, was the BodyMedia Fit, an armband whose data deviated from the lab-grade monitor's results by less than 1 percent. A distant second, at 11 percent off, was the Fitbit, a thumb-size device that clips onto your waistband.
@codewienie, nothing like hearing from the real user! Thanks for posting your observation. May I ask why you got so involved in trying out so many wearable devices? Was it for the purpose of hands-on product comparison for your work; or are you simply so health-conscious?
I use a Fitbit One, Withings Pulse, Striiv Play, Basis B1, Nike Fuelband, Jawbone UP24, Bodymedia Fit LINK, and the Omron USB. In my opinion, the UP24 and Bodymedia Fit LINK are the worst of the lot, in terms of accuracy. They record about 10% of what my Fitbit One, Withings Pulse, and Striiv Play report. Next worst is the Nike Fuelband. It's step-count is off nearly as much as the UP, however, its saving grace is the arbitrary use of "Fuel" as a measure of your activity. It's not base on anything scientific but the iphone application awards you "trophies" and pretty animations. My favorite is the Basis B1 because of it's transparent use. It records several variables to satisfy an algorithm that reports, in my opinion, a pretty accurate daily caloric burn and it records your sleep type, duration, and quality, without user intervention a la fitbit and withings.
An independent comparison of the numerous fit devices on the market today has revealed that their results differ by almost a factor 10. They may be great tools to motivate people, but they are not yet accurate enough.
I am glad that a voice as authoritative as that of the author from IMEC said it. I think we have always suspected that these things aren't that accurate, but wow, a factor of 10? Talk about a huge parity.
That said, as I read this piece, I suddenly realized one thing. How many people would like to wear a patch on their chest?
Yes, our health is important. And yet, I suspect the reason why many people wear those wearables today is not for the accuracy but for the looks. Is it not?
They want to let the world know that they are health conscious. Hiding that patch under your shirt would not serve that bragging purpose.
What are the engineering and design challenges in creating successful IoT devices? These devices are usually small, resource-constrained electronics designed to sense, collect, send, and/or interpret data. Some of the devices need to be smart enough to act upon data in real time, 24/7. Are the design challenges the same as with embedded systems, but with a little developer- and IT-skills added in? What do engineers need to know? Rick Merritt talks with two experts about the tools and best options for designing IoT devices in 2016. Specifically the guests will discuss sensors, security, and lessons from IoT deployments.