Sleep monitoring related offerings started surfacing in the consumer market couple of years ago. More recently EASYWAKEme, another European startup, has thrown it’s hat in the ring.
While reviewing Zeo and aXbo last year, I found myself wondering what was the need for having a bedside clock hardware, since most of that computing could be done in a smartphone. Seems like the crop of solutions that followed (e.g. WakeMate, Lark, Zeo Mobile) thought of the same. EASYWAKEme follows the pattern: wearable sensor (on wrist) that monitors movement as a proxy for sleep phase, provides intelligent wake-up alarm and longitudinal insight into your sleeping habits. Their ‘how it works‘ page explains it well.
Quick market analysis shows that price is going to be a very important factor since there is a wide range right now and consumers will tend to favor the cheapest. EASYWAKEme (£118), aXbo (starts at €179), SleepTracker ($149) are at the more expensive end compared to competitors like Lark (starts at $99), Zeo (mobile is $99), FitBit ($99), Jawbone Up ($99). WakeMate has already started the downward spiral at $59.
Activity sensors like FitBit, and more recently Jawbone Up are also chomping at the bit to include sleep monitoring as their value proposition. Cheaper solutions that are just smartphone apps are competition too. I actually bought the $0.99 SleepCycle iphone app out of curiosity (it had >20,000 reviews, most were positive) and ended up forgetting about it after a week. I now put it in the same category as placebo: it’ll work only if you already believe it does.
So the real question is whether body movement tracking and analysis is really a dependable way for extrapolating sleep phases. I still believe (based on scientific persuasion like this, and this) that it is. The ultimate hope is that these low-cost innovations can make their way back to help with serious issues like sleep disorders and psychiatric disorders. Current solutions like polysomnography are too cumbersome to be done outside of a clinical setting.
Disclosure: The generous folks at EASYWAKEme offered to send me a review unit, so I’ll be testing that hypothesis over the next few weeks. Look out for an update to this review in near future.
July 2012 Update: Used the unit several nights. Not scoring high on usability, mostly because of underlying BlackBerry OS inelegance. Overall, the concept seems to work. I woke up fresh most of the mornings I used it. Some gory details below:
- Alarm setup is a pain due to BB limitations. Very tricky to use the trackball to setup wake time. Can’t see the cursor lot of times. Aarrgh..
- If I don’t exit the app after pressing sleep, the BB stays on. More than once it ran out of juice by the morning.
- Buggy..even when on “silent” mode, the alarm tone still came on after 3 minutes
- There is no way to shut the alarm immediately on BB! Had to run out with it to avoid waking up my wife. Why can’t i shut the BB alarm with one keystroke or with the button on EASYWAKEme device?
- The device vibrations don’t stop until I hold the button down long enough to shut it off. Not sure if that is intentional design. But quite irritating in the morning…
- The BB’s final step of putting app in “sleep” mode is not intuitive or written well in manual. Also, that should be a button right next to the place where you set the alarm, and not a menu option.
- I would have liked it if shutting off the device or just pressing its button in some way shut off alarm on BB. Without that, the user is forced to hunt down their BB in morning to shut the damn alarm.
- Battery lasted only 2 days during my trial.
The concept of sensor-based connected devices that help consumers manage a healthy lifestyle is certainly gaining traction. Consider FitBit, Zeo, DirectLife, miCoach, BodyMedia, GreenGoose as examples. It was only a matter of time before a startup in this space decided to go with a watch-like form factor.
I first heard about MyBasis during a talk given by Bharat Vasan (founder?) at Bay Area Quantified Self gathering in June 2010. He referred to it as PulseTracer back then, and described its use for pulse monitoring. Based on the current description on their website, the product concept seems to have matured. Similar to BodyMedia, it now has 4 Sensors: Pulse, Temperature, Accelerometer, Skin Conductance (i.e. moisture). It is USB and bluetooth enabled; and comes with integrated social functionality (gaming, sharing, rewards, etc.). There seem to be smartphone, iPad and online applications that help provide analysis and understanding of the collected data.
MyBasis certainly has the concept nailed: A smart device with multiple sensors and long battery life in a familiar form-factor + Always-on and connected to desktop, mobile and online dashboards that simplify analysis of the aggregated raw data + Integrated social features that help make it sticky and viral. If they play it right, this can be a hit. Mainstream competition from products (like the lackluster Polar offerings) is uninteresting and hardly addictive.
But critics can say that there are always technophile early-adopters (like me) who crave anything that is novel and web-enabled. So before we get over-optimistic about the impending success of such devices, consider two important caveats.
First, it’s not about sophisticated monitoring or granular data. Whether they realize it or not, the key value proposition for an average end-customer is the personalized insight that results from it. Gathering 24X7 data from multiple sensors is great, but it’s all pointless if the user doesn’t understand the ultimate picture that results from all that data. Most users are interested in revelations into their health and lifestyle, not numbers. So the way MyBasis handles analysis is going to be critical. I was disappointed with MyZeo and FitBit for the very same reason. An interesting approach that MyBasis seems to have is the creation of a virtual pet that gives a quick proxy of your overall status. It may sound silly, but abstractions like these have shown promise in encouraging self-monitoring and positive behavior change in users (e.g. see UbiFit project at University of Washington).
Second, the technology and device needs to be so well-integrated with a user’s lifestyle that they essentially ‘disappear’. If someone needs to put a headband one (like for Zeo) or remember to find a USB cable and synchronize every week to prevent data loss, you can be assured that it’s not going to work out long term. As a species, we humans have remarkable lack of discipline even when it comes to things/habits that are good for us. So the offering needs to add minimal extra work and be seamless with your daily life. Neil Versel at MobiHealthNews calls it “passivity”. Example: FitBit gets a better grade than Zeo in this regard. I clip it onto my belt (almost sub-consciously now) every day and plop it on to the USB hub (always connected to my desktop) once in a couple of weeks. That’s it.
MyBasis is still in early beta, so my impression is based on what their website claims and not actual usage. I’ve signed up for preorder and will update this review when I get my hands on one. Seems like the cost is a one-time $199 for now. It would make a lot of sense to have a subscription-based model of some sort though. Also, it’d be great to see such solutions go beyond just wellness and be tailored for medical-grade serious conditions like diabetes, hypertension, etc. All clues indicate that such applications are not very far in the future.
More products designed around actigraphy keep surfacing. aXbo is a ‘Sleep Phase Alarm Clock’ by an Austrian company. The basic idea is to differentiate sleep phases (light, deep, REM) using body movements that are detected via a motion-sensor worn on the wrist. The bedside clock unit has other features like power naps, auto sleep melody fade, use for two people, smart alarm etc. A desktop ‘research’ application downloads the data and helps analyze it. So the overall service is similar to Zeo or Fitbit, except it looks like aXbo was perhaps first-to-market (2006 press release discusses launch).
Remote motion-sensor based alarm clocks are probably a nerdy gizmo for now. But it does provide a good use case of how everyday appliances are getting smarter and generating volumes of data. It’ll be very interesting to see how such data can be leveraged to further sleep disorder research- as I wrote in my Zeo post.
Note that there are other products out there that offer the same functionality: FitBit, WakeMate, SleepTracker, and the somewhat unique Zeo.
I wrote about Proactive Sleep a while back and wondered how it would be to combine a sensor to automatically record sleep pattern. Well, Zeo gets one step closer to that.
Zeo system consists of a wearable headband that measure brain’s natural electrical activity. Although their blog has a high-level explanation of how it works, my understanding is that its a single-channel EEG, which seems to be a reasonable way to do sleep analysis in healthy individuals. That data is wirelessly transmitted to a bedside display and stored on an SD card.
Algorithms based on proprietary logic churn out a personal sleep score (called ZQ) to quantify the type of sleep you get. The display unit looks like a bedside alarm clock and shows current and past 2 weeks worth of sleep analysis. Also has some smart alarm clock features like SmartWake alarm that wakes you up at the most suitable time within half-hour of set time. You could upload the data to an online sleep journal through the SD card. The website gives graphs, trends and the ability log other supplementary lifestyle data that can affect your sleep. All that for $249. An additional $100 would get lifetime access to a personal sleep coaching program, which includes regular assessments, goal tracking, email tips etc.
There is no question that Sleep Science is a serious, mature field. Zeo can find its place as a useful adjunct for plenty of sleep-related disorders that affect people who are otherwise healthy. It’s not an FDA approved 11 channel medical grade polysomnogram, and it’d be a mistake to compare it to one. It’s perhaps a closer analog to Actigraphy where a wearable sensor measures overall motor activity during sleep. An actigraph unit is an accelerometer based device like the FitBit, WakeMate, SleepTracker or Axbo.
Accurate or not, Zeo is yet another proof that healthcare is slowly being transformed by sensor-based, portable devices that are capable of analyzing data in a consumer-oriented way to enable individual patient to manage their conditions better.
April ’10 Update: I ordered a Zeo from Amazon in March’10. Been using it for couple of weeks now. The hardware looks and performs well, but I have my complaints. The sensor in headband is bit bulky to sleep with, and I don’t think I’ll get used to the headband anytime soon. The fact that I need to replace the sensor pad every 90 days feels like a hidden cost (they are $24.95 for a pair without S&H) that was not disclosed completely in the marketing.
Software-wise, it’s bit less intuitive than I expected. I’m still wondering what the data tells me. I know Z-score is a overall indicator, but what about the breakdown? Are my 49 minutes of deep sleep enough? The fact that I woke up 7 times – is that normal? In the end, I’m asking the question- what does this $250 device do for me… Probably not much because I’m not the target audience for it. People suffering from sleep disorders would perhaps be more receptive.
May’10 Update: Returned it. Sorry Zeo.
June’10 Update: Learned about the Zeo API. Mobile fitness companies RunKeeper and Daily Burn are the first to integrate their services with Zeo. Neat strategy to promote it as a platform.
July’10 Update: Found out about NeuroVigil: another startup in the sleep-monitoring space. Their technology is also based on a single-channel EEG wireless sensor. They claim to be able to send the data (and its analysis) to any mobile device. NeuroVigil’s founder, Philip Low gave an interesting talk at TEDMED in 2009.
Nov’11 Update: Noticed that Zeo now has a cheaper, smartphone-based mobile ($99) solution and original clock has been re-packaged as bedside solution ($149). Smart move, since it makes them more relevant in comparison to other solutions like EASYWAKEme the market.
Fitbit is a startup trying to leverage the rising wellness trend- they manufacture a small device that you can wear throughout the day to find out how much physical activity you did, how well you slept etc. The tiny device sells for $99 and contains a motion sensor (like the Nintendo Wii) that converts 3D motion data into useful interpretations about your lifestyle. In a 7-day period, whenever you are in range of it’s base-station, the data is automatically uploaded to the Fitbit website where you can get detailed view into the past data and summarized reports. There are also some social networking aspects built in (like collaborate on fitness goals with your co-workers, friends)The tracker measures the intensity and duration of your physical activities, calories burned, steps taken, distance traveled, how long it took you to fall asleep, the number of times you woke up throughout the night and how long you were actually asleep vs just lying in bed. Regardless of data accuracy and clinical efficacy, I think the concept is brilliant. We are slowly moving away from the I’ll-use-what-my-insurance-pays-for mentality to a more proactive mindset where consumers are taking thier wellness matters into thier own hands.
Fitbit represents the impending wave of consumer-oriented, internet enabled service providers that effectively combine hardware and software to provide a compelling solution around a healthcare issue. The future holds great potential for similar lifestyle-management solutions that can help affected individuals manage their chronic diseases (heart failure, diabetes, hypertension, asthma, etc.)
18th December 2009 Update: Got my FitBit in mail today! Will use it over next few months and report back.
October 2010 Update: Still using it. Satisfied with the battery performance, easy of use. Unsatisfied with the analytics and interpretation provided. They now charge $49 for a premium account that gives in-depth analysis of your data. I’ve not signed up yet. Google Health partnership was announced recently too. Fitbit will share 3 things: calories burned, steps taken and daily distance with Google Health.