I’ve now managed to read through all 11 pages. The Systems Engineer in me obsessively wants to structure the thread into topic areas, but I now feel I understand the limits of what is and might be possible with the Mobile.
I have the diagnostics sqllite file opened in SQL Lite Database Browser and can see all the fields discussed so will have to find time to try exporting it to Excel and playing with some graphing functions. I work with some Software Apprentices - maybe I can interest them in playing with the data, see how far we can get!
When I thought I wasn’t going to be able to get a Zeo, I bought a Jawbone UP, thinking it would be almost as good. Hmmmm. No comparison to the Zeo. For example, it doesn’t recognise the difference between REM and deep sleep, so lists it all as ‘Deep Sleep’. Also, as I’m running both the UP and Zeo in parallel, the Zeo typically records me waking 3 to 7 times per night, whereas the UP rarely records me waking anything more than twice. Anyway, one of the things I wanted to check is activity versus sleep, to see if there is any correlation. I’d really like to look at activity data from the UP versus sleep from the Zeo. As people have been talking about heart rate vs sleep, I just wondered, is anybody looking at activity/steps versus sleep combined from different devices? Or extracting data for localised processing from the UP, in general? (please feel free to point me at a better thread, if one already exists!).
Jane
Hi
Oh, and I meant to also say, as I was never enrolled on the zeo website, nor used the zeo personalised advice services, as I didn’t get my zeo until after they closed, can somebody advise what data/information could you only get via the website? (I want to know what I’ve missed ). I know my GP says that % of sleep/total time in bed is used for clinical determination of insomnia, but what other figures might be useful if you wanted to present the data to a sleep expert clinician at some point down the line? Or are there good websites that discuss those issues that somebody could recommend? Those should hopeuflly be relatively simple to calcluate from the extracted data, even though not in the current Apps.
Jane, I don’t think you missed too much by not having access to the website. Most of the advice given to improve sleep was basic sleep hygiene techniques which can be found through Google. Their site was nicely done and one could observe personal trends in various aspects of sleep as well as comparisons to norm without spending a lot of time doing spread sheet work. However, my guess is the more analytically inclined bedside users exported the data and did their own spreadsheet/chart work, anyway.
In my opinion, the best way to determine the quality of your sleep is to have a professional polysomnogram done, and is probably warranted if you are suffering from unexplained excessive daytime sleepiness or fatigue. While the Zeo can provide some insight and is, I think, a great tracking tool, it can not tell you, for example, if you are suffering micro arousals or PLMD.
Yeah, read plenty of the ‘sleep hygeine’ stuff before and have tried most of it and found it mostly unhelpful!
I periodically have really bad sleep, mainly because I have such an erratic work schedule, but my ‘average’ sleep has improved since I’ve been monitoring it. I think it’s the opposite of being pychosomatic!!
Since I was diagnosed with high blood pressure earlier in the year and bought a iPhone-driven blood pressure monitor, I decided that I should turn my body into my own science experiement rather than feeling like I was a guinea pig for the medical profession. I’m having great fun with it!
The android app doesn’t appear to have a way to email the data for diagnostics. I’m guessing that method only works with iOS. However, it’s still possible to use adb from the android development kit to back up the data. Get the sdk from http://developer.android.com/sdk/index.html and run ‘adb backup com.myzeo.android’ Then you can use the android backup extractor from http://sourceforge.net/projects/adbextractor/ to get the data out of the backup. You wind up with a zeo.db sqlite file.
Having fun with my data set. Over lunch at work, I’ve created some basic charts in excel to examine how time in each sleep phase and % sleep in each phase varies with day of the week and date. Next to see how that varies by time that I start sleeping.
I did notice that the diagnostics file is quite big - 7MB. Does anybody know if that is because of the database structure, so the size shouldn’t change much with number of days of data, or, given I had only 10 days of data in that set, that file is going to get enormous! As my only method for data transfer currently is email, I’d like to cut that file size down, if it’s the latter. Is there a method to delete all the stored data/reset the app? I can’t even see an obvious way to delete individual days of (unsynced) data, let alone the whole dataset! Otherwise, I guess it’s the apps that let you see the iPhone file structure and hope it doesn’t ‘break’ the app!
Have found out that one of the guys in our office runs a home automation app for remote temperature and power monitoring on a Raspberry Pi that might be easily modifiable to take data over bluetooth. I’ll keep you posted if that works out.
If you’re still around, I’d be very interested in this spreadsheet. I just started producing my own today, over lunch, as I don’t have much spare time. Looks like you are far more advanced than me, so I’m all for reuse wherever possible!!
Jane
Hi zeoleth
Did you get anywhere with the Pi? I have one and would very much like to use it for data logging and processing from my Zeo mobile headband as the iOS app I use now is far from ideal!!
I am not sure why your sqlite file is that big. When I installed the app the initial database file (ZeoDataStore_v1.1.sqlite) was only 45 KB.
I added my own data (from my bedside Zeo) to the file and then pushed it back to my iPad and then was able to view the bedside data using the mobile app. 2 nights of sleep and the file size was still around 45 KB, but I did not populate all of the diagnostic fields only the ZZOSLEEPRECORD table.
Sorry, I should have been more specific, the whole email that my iPhone/iPad sends is 7MB, and, as you say, the data store file is only 1MB of that, but the iOS also sends an XML data representation - .plist - which is 6MB. Even the data strings for each message in the plist are very different to those in the data store, so I’m assuming I can’t decode it as simply as the data store. Hence, I probably don’t really need it, but the file size is huge.
Jane - thought you might enjoy this graphic: http://www.cpaptalk.com/viewtopic.php?f=1&t=85888&p=781664#p781664
One can argue the validity, but it is interesting. After reading a recent article on the impact of moon phases on sleep, I’d like to chart that, but I haven’t gotten around to it.
Jay
For those that haven’t looked, it’s data showing a correlation between % deep sleep versus temperature.
I absolutely noticed a correlation between temperature and how I feel that I’ve slept - I tend to keep my bedroom very cool as a result. And not wanting to alienate half the population, but at certain points in the menstrual cycle, hormonal levels affects women’s ability to regulate their temperature and, personally, my sleep always suffers, so that also supports the same data. Making me feel like this idea of adapting a home automation logging system, that was reading data from room temperature sensors, might be even more of a good idea!! Can you get bluetooth or wifi skin thermometers??
It was also interesting for a different reason - I was planning to google to find out what is a ‘normal’ % deep sleep. Those graphs top out at 16%. Although I have an occasional bad night, I’m typically >12% or approx 1 hour of deep sleep. Seems that is pretty good in comparison to that data!
I haven’t tried either, but the data collection and analysis with Zeo data could be fun.
There is also the medication approach to change sleep structure, but then the question is do the benefits outweigh the risks. You can certainly see the correlation between dose and % Deep sleep on the Zeo results.
That’s what to miss about the Zeo website. They provided great reference material. From Zeo and elsewhere, “typical” deep sleep comprises about 21% of sleep in those adults in their 20’s and declines steadily to less than 10% in people aged 70 and older. Or stated another way, about an hour and half in young adults to less than 30 minutes in the elderly. REM remains relatively constant with age at about 22% or about an hour and half declining to a little more than an hour with age. According to Zeo*, “typical” for someone in their 30’s would be 7.1 hours total sleep, with 1.5 hours REM, 69 minutes Deep, interspersed by 21 minutes wake. Zeo’s average ZQ for a 30 something is around 80.
*“How does my Sleep Compare”, provided by Zeo with bedside model
Thanks Jay! Being in my mid-40’s, assuming it’s a straight-line graph, I’m spot on with approx 1 hour of deep sleep then! Probably even getting more REM than average, as typically nearer to 2 hours, and less overall wake time, though often split across several different events. Very interesting!
Jane
Hi
I can’t see it explicitly mentioned in prior posts, but that maybe because I don’t fully understand some of the previous documents, but perhaps someone can point me in the right direction if that is the case…
Do we know what is the data that is transmitted over bluetooth, format, content, any encryption? Is there pre-processing in the sensor you wear on the headband, or is all the processing and inference about sleep state in either the bedside device or iphone/iPad/Android app? Does that vary between mobile and bedside versions?
If I’m going to use the data logging approach, and only have the mobile version, it would be simplest if I could do that in real-time and connect my headband directly to the raspberry Pi, over bluetooth, however, if the proprietary algorithms that convert braiwave data to inferred sleep state are only in the app, I will have to manually extract the data from the app and process it offline later. If I’m just looking at things like temperature, that vary relatively slowly, time synching betwen datasets isn’t likely to be critical, but, you know, if I can do it all at the same time, it would be less hassle and therefore better
I am new to the Zeo world. I plan to purchase one of the Zeo systems and I am now doing research before I buy. I have seen three basic systems. The bedside, the mobile, and the mobile Pro. It appears as though Ebay is the best option. I would appreciate help in selecting the best of the Zeo systems for home use. I have downloaded the update files for the firmware and the data, and it appears “doable.” Feel free to email or post. Any and all help will be appreciated.
I bought a second hand Zeo Sleep Manager Mobile, which is on its way. I wanted to prepare my Iphone 4s for it, so I looked for the Zeo App in App Store, but nothing that I found.
Is there any way to use Zeo Sleep Manager with my iPhone?
Having checked some posts in the forum I understood that if I had the app I could export the data, but found no reference to those without the Zeo App installed.
Please help me with a hint
Thanks a bunch
In the meantime I spotted a post referring to some .ipa file, which must be the installer of apps on iPhones. So my question is where I can find a Zeo Sleep .ipa running on my Iphone4s?
Ron, in my opinion, the “best” Zeo system is the Bedside, though it is probably the most expensive. The Bedside model displays a graphical hypnogram of the previous night’s sleep and allows review of a 7 day history. The “ZeoDecoderViewer” application allows one to download and view any previous night’s hypnogram and the “ZeoDecoderViewer” and “ZeoDataDecoder” applications allow access to all historical data in a CSV type format. This historical data can be transferred to a spreadsheet and charted as appropriate to illustrate trends, etc. in a fairly straightforward fashion. Additionally, with a little hacking, the Bedside model can be used to monitor sleep and or brainwaves in real time. See post #156 for these apps.
To be fair, I don’t own a mobile device, so my opinion of their usefulness is based solely on posts I have read. But it does seem people have quite a bit of trouble getting them set up and then getting much more than a graphical representation of the the previous night’s sleep. Historical data seems to be very difficult to obtain with mobile systems.
Again, my opinion of mobile systems is based only on posts I have read. Maybe mobile users can offer more insight as to the mobile capabilities and methods for historical data retrieval?
@Tisztul_A_Visztula - the Zeo app was presumably removed from the app store because it is not compatible with iOS 7 and, as Zeo have gone out of business, there is nobody to update the app. Although some of us have the installation file, I understand it’s configured only to work with the purchasing iTunes account, so you wouldn’t be able to use it, even if we sent it to you. Sorry!
@Ron - I only have experience of the Zeo mobile with the iOS app.
It is still possible to email to yourself your daily data as a graphic file, (.png) from the iPhone/iPad app
It is also possible to export all your historic data, using the ‘diagnostics’ option described earlier in the thread. This exports the data as an sqllite database and XML. This can be exported to csv and imported into something like excel to manipulate the data, however, note that the apps that have been developed by others for use on data from the bedside model, they DO NOT work with the exported data from the mobile system, because the data pre-processing is different, i.e. the fields don’t match. (You may be able to modify that manually). The data is not the most convenient to manipulate, straight out of the mobile system, so, for example, my colleagues and I still haven’t managed to get an excel macro working exactly as hoped, to recreate the hypnogram that the app produces.
I believe there is more scope to write apps to interface with the Android Zeo app, which is still available on Google Play, but my Android phone keeps dropping the bluetooth connection with the headset, so I’ve not had chance to do much with that yet!
So, essentially, unless you have the skills or have friends with the skills to help with the data processing side, or only want to view the trends graphically from the output png files, the mobile is the harder work option, but therefore the much cheaper option to buy secondhand
In any case let me give a try. Could you please send the installation file to ujeletmgtsz at gmail.com ?
Thanks,
Zoltan
On the other han can somebody parallelly tell me which non-Zeo iPhone app is the best to use Zeo Sleep mobile as a paired bluetooth device? I guess there must be some third party apps which can display the data collected from Zeo mobile.
). I know my GP says that % of sleep/total time in bed is used for clinical determination of insomnia, but what other figures might be useful if you wanted to present the data to a sleep expert clinician at some point down the line? Or are there good websites that discuss those issues that somebody could recommend? Those should hopeuflly be relatively simple to calcluate from the extracted data, even though not in the current Apps.