I saw a talk on consumer wearables in sleep at the Sleep Trends conference this spring. There was a great presentation comparing consumer sleep trackers, and how their advertised reporting of "accuracy" can be misleading.
If you do a thought experiment: let's say I make a "sleep tracker" that is just a piece of plastic, no sensor, and write an algorithm that always presumes the wearer is in stage N2 sleep. Since we spend around 50% of our night in N2 sleep, this algorithm would start with a fairly high baseline accuracy of 50%.
Sensitivity and specificity are much better measurements of accuracy, because they measure the false positive and false negative rates. A comparison of Oura against a medical polysomnogram showed that while it's very good at detecting sleep, it's very bad at detecting not sleep:
From https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6095823/:
"ŌURA ring had a 96% sensitivity to detect sleep, and agreement of 65%, 51%, and 61%, in detecting “light sleep” (N1+N2), “deep sleep” (N3),and REM sleep, respectively. Specificity in detecting wake was 48%"
The current state of the art in sleep measurement really requires some sort of EEG measurement beyond just PPG and HR, what Oura uses. The Dreem headband is a good example (https://dreem.com/), but it's definitely more cumbersome than a ring. It's tailored more like a medical service for insomnia than for the average consumer interested in quantified self.
I own a Dreem V2 headband and though it's more cumbersome than a ring, it's really not a problem. It takes me less than a minute to put it on and activate it each night and though it's a little uncomfortable at first, I got used to it quite quickly.
I disagree that it's "more like a medical service for insomnia". The app does have some of your standard CBT stuff in it if you choose to use it but there's also a lot for people without major sleep problems as well. For example there's a sleep schedule function that will give you bedtime reminders and automatically wake you in the morning when you're in a light stage of sleep.
As far as data, I'd say that's one of the major focuses of the app. "Sleep reports" (weekly summaries of your sleep habits) are made very prominent in the main screen of the app and contain reasonably comprehensive but very approachable pieces of data.
As for accuracy, I've compared various products of various kinds such as under bed, apps and wristbands (though not Oura) and found basically everything but Dreem to be pretty garbage at doing anything more than detecting that I'm laying in a bed. I'm not convinced it's possible to detect that I'm awake but lazily delaying leaving the bed with anything but an EEG.
Garmin acquired FirstBeat Analytics less than two weeks ago [1]. According to DC Rainmaker, "They power the vast majority of Garmin’s fitness and outdoor training-focused algorithms" but they are also used by other manufacturers. The Garmin's Sleep Stage tracking, powered by FirstBeat Analytics [2], use higher power red LEDs at night. I wonder if OURA uses a separate technique, maybe even transmissive PPG.
>It's tailored more like a medical service for insomnia than for the average consumer interested in quantified self.
There used to be a EEG headband aimed at the quantified self group (cost $100 or $150) but they went out of business. Probably because for that group a fitbit like tracker was good enough. So I'm guessing the medical side and higher price point is seen as a better business model.
Ultra-low power indeed: I often get 6+ days of use before having to recharge. I’ve been very happy with mine; the sleep tracking seems quite consistent with my subjective experience, and the activity/exercise detection matches closely with the data from my Apple Watch.
I love the aura ring. AFAIK it is clearly the best non-intrusive sleep tracker and I would recommend it to anyone for sleep tracking. But in my experience the activity tracking is horrible because the ring achieves low power consumption by being off most of the time. I would suggest using a different device for daytime activity tracking.
I have an Apple Watch that I wear overnight and have a sleep tracking app. Have you compared your ring to your watch to see if they report similar results? I have been considering the ring for HRV tracking (the Apple Watch only tracks HRV at random times when you are still, whereas the ring apparently tracks continuously) but also wondered if the sleep tracking is superior. Thoughts?
Before buying one I tried to make research about its effectiveness and there are some very alarming reports, coming, of all places, from the research published by the same company. So this isn't even biased...
“From EBE analysis, ŌURA ring had a 96% sensitivity to detect sleep, and agreement of 65%, 51%, and 61%, in detecting “light sleep” (N1), “deep sleep” (N2 + N3), and REM sleep, respectively. Specificity in detecting wake was 48%.”
Specificity in detecting wake was 48%! If this was a medical test, it would never be approved by FDA.
A specificity of 48% means that there is a 48% chance that someone is awake when the device says they are asleep.
That is horrible.
This is very surprising and alarming. As the author later goes on to describe, this has real implications and possible negative effects, this is not just a benign error here and there. Users are apparently supposed to use the Oura data to change or at least adjust and improve their sleep habits, and of course if they will do it based on faulty information, the adjustments themselves are going to be faulty and it can lead to worse or sub-optimal sleep!
Apparently you can still use Oura to track heart rate and HRV, but their own proprietary markers like “readiness” are probably based on all of their data including sleep, so they are not going be that accurate either.
This matches my experience having received one Nov 2018 (and returning it a few weeks later). It was constantly getting my actual sleep times wrong, more often than not marking me as being asleep while watching tv in the evening.
After contacting support, their solution was an upcoming update to their app where I'd be able to edit the data so that I could override the app whenever I knew it was wrong. Which completely invalidates the primary reason to own this product. I mean, if I knew when I was asleep, why would I need a ring to track it?!
So given that it couldn't properly track sleep, doesn't track activity (by design), the only other purpose in my mind was to track HRV. And count me as skeptical of the accuracy of that data as well.
I think the idea is similar to what the NBA is doing: monitor doctors and nurses with the ring to try to detect COVID. But they are trying to properly study the idea.
I am impressed with my ring after having had it for about a year. The battery life seems to be as good as ever. The only concern I have is that I often have to wear it on a different finger. My hands are less bloated, which I credit partly to having been more active since I’ve had it.
From my experience the last thing you need for good sleep is a tracker. I’d try to fix sleep hygiene first and adapt a strict sleep schedule you do every day.
Some people (myself included) fix (sleep) hygiene problems by first closely watching and diagnosing them.
Your comment would sound silly if it said "last thing you need for good code is a debugger, I'd try to add a linter first". Or "last thing you need to maintain a good weight is a scale, I'd try to eat better first".
Neither are exclusive, and the former helps the latter if you use it well.
Yeah this seems like a solution in search of a problem. I don't see how it's worth the price. Reading the comments in this thread gives me an interesting view into why people choose them.
If there is a primary flaw, it's that it can't be resized. If you are sufficiently motivated to become more active, the ring becomes too loose over time.
Not the worst problem, but it ends up in caught in the drain more than I'd like.
I wonder if it's currently technically feasible to make a version with swappable batteries while maintaining the same form factor, functionality and other properties the product has as-is.
I know a lot of other products could have better serviceability and repairability if one were to sacrifice form factor, dimensions or aesthetics, but for a product like this it doesn't seem obvious to me how one would go about that, but it'd be very interesting to speculate on.
It achieves low power consumption by being off most of the time. That works out just fine for sleep because sleep phases take a long time to change and to verify that you need to take multiple samples over long time periods. But to accurately track movement characteristics you need to be always on.
The instructions it came with tell you not to wear it while using dumbbells or barbells. As for cardio, I guess it's ok. Mine tends to say I took 20% more steps per day than reported by the pedometer app on my phone. I have no clue which one is more accurate though.
Guessing the extremely tiny battery has something to do with it. It's a bit like how Apple Watch's battery life tanks when you use it for exercising and I don't think the Apple Watch gives the most accurate/highest resolution readings.
I would assume that to achieve a 7 day battery life they only take rare measurements of your heart rate, which makes it unfit for the kind of high resolution heart rate tracking people are used to from fitness watches. They also mention that any sport which has a significant level of hand involvement might disturb the reading.
They do still track general intensity of activity though, which for general health is probably a more relevant metric than exact heart rate.
[+] [-] z3ugma|5 years ago|reply
If you do a thought experiment: let's say I make a "sleep tracker" that is just a piece of plastic, no sensor, and write an algorithm that always presumes the wearer is in stage N2 sleep. Since we spend around 50% of our night in N2 sleep, this algorithm would start with a fairly high baseline accuracy of 50%.
Sensitivity and specificity are much better measurements of accuracy, because they measure the false positive and false negative rates. A comparison of Oura against a medical polysomnogram showed that while it's very good at detecting sleep, it's very bad at detecting not sleep:
From https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6095823/: "ŌURA ring had a 96% sensitivity to detect sleep, and agreement of 65%, 51%, and 61%, in detecting “light sleep” (N1+N2), “deep sleep” (N3),and REM sleep, respectively. Specificity in detecting wake was 48%"
The current state of the art in sleep measurement really requires some sort of EEG measurement beyond just PPG and HR, what Oura uses. The Dreem headband is a good example (https://dreem.com/), but it's definitely more cumbersome than a ring. It's tailored more like a medical service for insomnia than for the average consumer interested in quantified self.
[+] [-] Youden|5 years ago|reply
I disagree that it's "more like a medical service for insomnia". The app does have some of your standard CBT stuff in it if you choose to use it but there's also a lot for people without major sleep problems as well. For example there's a sleep schedule function that will give you bedtime reminders and automatically wake you in the morning when you're in a light stage of sleep.
As far as data, I'd say that's one of the major focuses of the app. "Sleep reports" (weekly summaries of your sleep habits) are made very prominent in the main screen of the app and contain reasonably comprehensive but very approachable pieces of data.
As for accuracy, I've compared various products of various kinds such as under bed, apps and wristbands (though not Oura) and found basically everything but Dreem to be pretty garbage at doing anything more than detecting that I'm laying in a bed. I'm not convinced it's possible to detect that I'm awake but lazily delaying leaving the bed with anything but an EEG.
[+] [-] sradman|5 years ago|reply
[1] https://www.dcrainmaker.com/2020/06/garmin-acquires-firstbea...
[2] https://assets.firstbeat.com/firstbeat/uploads/2019/11/First...
[+] [-] marcinzm|5 years ago|reply
There used to be a EEG headband aimed at the quantified self group (cost $100 or $150) but they went out of business. Probably because for that group a fitbit like tracker was good enough. So I'm guessing the medical side and higher price point is seen as a better business model.
[+] [-] dirtyaura|5 years ago|reply
Note: If you are interested in working at Oura, we are hiring SW Engineers both in Helsinki, Finland and San Francisco.
Current open positions can be found on https://ouraring.com/careers and new positions are opening in a few weeks.
[+] [-] robterrell|5 years ago|reply
[+] [-] amacbride|5 years ago|reply
[+] [-] gregwebs|5 years ago|reply
[+] [-] voisin|5 years ago|reply
[+] [-] Hamuko|5 years ago|reply
[+] [-] trollied|5 years ago|reply
[+] [-] Erlich_Bachman|5 years ago|reply
Apparently you can still use Oura to track heart rate and HRV, but their own proprietary markers like “readiness” are probably based on all of their data including sleep, so they are not going be that accurate either.
[+] [-] wvl|5 years ago|reply
After contacting support, their solution was an upcoming update to their app where I'd be able to edit the data so that I could override the app whenever I knew it was wrong. Which completely invalidates the primary reason to own this product. I mean, if I knew when I was asleep, why would I need a ring to track it?!
So given that it couldn't properly track sleep, doesn't track activity (by design), the only other purpose in my mind was to track HRV. And count me as skeptical of the accuracy of that data as well.
[+] [-] Jhsto|5 years ago|reply
Further, even though Oura is a private company, its financial information is public (as is any Finnish company's): https://www.finder.fi/Tuotekehitys+tutkimus-+ja+suunnittelup...
[+] [-] dirtyaura|5 years ago|reply
[+] [-] BiteCode_dev|5 years ago|reply
I misplaced it in the washing machine twice and it still works as brand new.
Besides the proprietary aspect of the platform and false positive reporting sleeping while not, I have only good things to say about mine.
Nifty little device.
[+] [-] gregwebs|5 years ago|reply
I think the idea is similar to what the NBA is doing: monitor doctors and nurses with the ring to try to detect COVID. But they are trying to properly study the idea.
[+] [-] krustyburger|5 years ago|reply
[+] [-] wazHFsRy|5 years ago|reply
[+] [-] scrollaway|5 years ago|reply
Your comment would sound silly if it said "last thing you need for good code is a debugger, I'd try to add a linter first". Or "last thing you need to maintain a good weight is a scale, I'd try to eat better first".
Neither are exclusive, and the former helps the latter if you use it well.
[+] [-] theshrike79|5 years ago|reply
Do I sleep better when it's colder, how about different pillows, blankets? What does alcohol do to my sleep? etc.
With the ring I've got actual data to back up a general feeling of did I sleep well/badly.
[+] [-] crmrc114|5 years ago|reply
[+] [-] underyx|5 years ago|reply
Wait, what? I got mine in Feb 2019. Did they re-release it or something?
[+] [-] dirtyaura|5 years ago|reply
[+] [-] social_quotient|5 years ago|reply
[+] [-] unicornporn|5 years ago|reply
[+] [-] dirtyaura|5 years ago|reply
https://iopscience.iop.org/article/10.1088/1361-6579/ab840a
https://www.researchgate.net/publication/335764849_Following...
[+] [-] sadfasdfsad|5 years ago|reply
If there is a primary flaw, it's that it can't be resized. If you are sufficiently motivated to become more active, the ring becomes too loose over time.
Not the worst problem, but it ends up in caught in the drain more than I'd like.
[+] [-] oneplane|5 years ago|reply
I know a lot of other products could have better serviceability and repairability if one were to sacrifice form factor, dimensions or aesthetics, but for a product like this it doesn't seem obvious to me how one would go about that, but it'd be very interesting to speculate on.
[+] [-] theshrike79|5 years ago|reply
Any kind of swappability would require connections, which in turn would compromise the IP rating.
[+] [-] cinntaile|5 years ago|reply
[+] [-] gregwebs|5 years ago|reply
[+] [-] swebs|5 years ago|reply
[+] [-] Hamuko|5 years ago|reply
[+] [-] hobofan|5 years ago|reply
They do still track general intensity of activity though, which for general health is probably a more relevant metric than exact heart rate.
[+] [-] xnzakg|5 years ago|reply
[+] [-] nwah1|5 years ago|reply