Extracting Snippet of Data from Sleepyhead

I had kidney stone surgery about 3 1/2 years ago & several procedures since. None have been fun. My doctor always asks if I wake up at night to go to the bathroom. Since I've had CPAP (almost 2 years now) I get up much less often at night, but still 50-60% of nights. Sometimes, multiple times. I've decided that having to pee isn't always what wakes me up. Instead, I think it's often an apnea "event". I usually have to (& do) go, but was it enough to wake me up?

I don't understand Sleepyhead well enough to interpret all the charts. Even if I did, reviewing a day-at-a-time for more than a few days would be tedious, at best. Is it possible to analyze SH data (in "batch mode") & flag by date/time (for instance) when an apnea event occurred just prior to my turning CPAP off for a bathroom break. I guess what I'm asking is whether SH will produce customized reports. Or is the SD card data in a format that can be processed by other software?

This information would be very helpful to my doctor (& me) in evaluating my kidney problems, which are much more intrusive (& painful) than apnea & CPAP.

You can export data from Sleepyhead. Look at the export data function under the file menu. You'll want to choose "CVS file export". It's fairly self explanatory, but if you need help, let me know. For what you want, probably best to pick a custom data range, and "details". This will give you a comma separate values file (CSV), that can be easily imported into any spreadsheet application. A number of event types are exported, along with the date and time stamp of the event, and pressure at that moment.

linuxman wrote:You can export data from Sleepyhead. Look at the export data function under the file menu. You'll want to choose "CVS file export". It's fairly self explanatory,

it's also pretty useless. just good for day to day trending. no detailed data.

Robear wrote: reviewing a day-at-a-time for more than a few days would be tedious

What are you trying to do that seems to be tedious?

As I understand, you are looking to see if there is an apnea(s) just before you turn the machine off to get up for Pinkelpause. This is fairly easy to do once you have your charts formatted - https://sleep.tnet.com/resources/sleepyhead/shorganize.

Then you just flip through the daily reports and look for the breaks in the chart where you took Pinkelpausen. Then just look at the events line in the minutes before the break.

Depending on your work schedule, you could do this once each day or once on the weekend to cover 7 days.

In either case, this is quick and easy once you have everything set up properly.

Am I understanding what you are trying to do?

palerider wrote:

linuxman wrote:You can export data from Sleepyhead. Look at the export data function under the file menu. You'll want to choose "CVS file export". It's fairly self explanatory,

it's also pretty useless. just good for day to day trending. no detailed data.

Yeah I know.. I've used it. He doesn't seem to be looking for detailed data, however. He just wants events, and it does export events with a timestamp using the details setting. Each power cycle will create a new session, and the session number is also listed in a separate column. It will be very easy to spot events near those boundaries. Of course you can do the same in SH pretty easily, but the spreadsheet probably would be a bit easier to scan through for multiple days.

linuxman wrote:

palerider wrote:

linuxman wrote:You can export data from Sleepyhead. Look at the export data function under the file menu. You'll want to choose "CVS file export". It's fairly self explanatory,

it's also pretty useless. just good for day to day trending. no detailed data.

Yeah I know.. I've used it. He doesn't seem to be looking for detailed data, however. He just wants events, and it does export events with a timestamp using the details setting. Each power cycle will create a new session, and the session number is also listed in a separate column. It will be very easy to spot events near those boundaries. Of course you can do the same in SH pretty easily, but the spreadsheet probably would be a bit easier to scan through for multiple days.

fair 'nuff.

I exported both Sessions & Detail data (2 exports). I selected columns & combined into a single spreadsheet using the Detail DateTime & the Session End-DateTime. Then I sorted on the DateTime column.

That yields:
----- DateTime ------ Session --- Event
2017-02-11T03:54:45 1837 Pressure
2017-02-11T03:58:13 1837 Hypopnea
2017-02-11T03:59:13 1837 ClearAirway
2017-02-11T03:59:19 1837 PressurePulse
2017-02-11T04:02:50 1837 VSnore2
2017-02-11T04:03:19 1837 Pressure
2017-02-11T04:06:50 1837 VSnore2
2017-02-11T04:07:01 1837 Pressure
2017-02-11T04:07:01 1837 Session End
2017-02-11T04:10:03 1838 Pressure
2017-02-11T04:12:03 1838 VSnore2
2017-02-11T04:13:16 1838 RERA
2017-02-11T04:13:45 1838 VSnore
2017-02-11T04:13:45 1838 Pressure
etc
This example seems to me to show that events, not blatter pressure, is what woke me up. A little more work should enable quick locating & reviewing events immediately prior of Session Ends. And I need to learn more about the various types of events.

I'm curious why there were occasional gaps in session#s, usually between morning & night sessions.

Robear.
I appreciate what you are trying to do, but I don't know that you will be successful. I think you will find evidence of "events" close to nearly all "session ends" and probably session starts as well. Here's the problem: As we become aroused and enter lighter stages of sleep our breathing becomes irregular with pauses between breaths. The machine detects these pauses, doesn't know that we are awake or wakening, and scores an apnea, usually, but not always, a Central Apnea. These scored events are often referred to as "Sleep-Wake Junk" meaning they wouldn't count in a laboratory setting. Disturbances from any cause, bladder included, are likely to rouse one from sleep producing irregular breathing patterns which then might be scored as events.
Here's a snip from my own flow rate graphic of last night near a session end where I indeed had to get up to pee. Notice that around 04:47 my breathing becomes irregular signifying a likely disturbance or change in sleep stage. Then around 04:47:20, an event, a UF1* is scored, followed by a 2nd and then a CA around 04:48:15. Compare the waveform before 04:47 to that after and see that it becomes irregular after and remains so until I shut off the machine at 04:49:00.
In this example, what looks to be a disturbance around 04:47 is followed by several events and culminates in a wake and end-of-session to pee. Was the initial disturbance caused by the bladder or something else? I have no way to tell. It could have been a shift to REM resulting in arousal where I became aware of a full bladder. Or, it could have been a signal from the bladder, initially. I don't know and I don't know of any way to tell using our standard home equipment.


*(UF1 is a preferential setting of Sleepyhead which can be used to score events other than the standard 10sec apneas scored by the machine. In this example, UF1 depicts apneas from 5-10 seconds in length.)

Robear wrote: 50-60% of nights. Sometimes, multiple times.

How much volume?

One trick is to drink a glass of water at 5:00 p.m. and then nothing the rest of the evening.

Make sure to drink sufficient fluids at breakfast and lunch. We need to keep hydrated.

I agree with Jay that it's going to be tough to differentiate what actually causes the events near those boundaries. It might be possible with a bit of signal processing to make some inferences looking at data when you do and do not get up to pee..but gotta have the second by second flow data for that. BTW, it's on my todo list to work up some changes for Sleepyhead to export all data so one could do their own post processing, when I get time. Not sure if the author would accept the code, but it's GPL, so I can go on my own if I wanted to...

Thanks, Jay & Linuxman, for you input. You are right. There's just not enough info to draw firm conclusions. But I just had to look for myself. I must admit (at the risk of provoking "assaults" from some) I don't import into SH frequently. I enter transmitted data into a spreadsheet and compute averages, etc as flags for possible problems. That gives me a concise, quick-look overview. Fortunately, I haven't had major problems for over a year now.

However . . .
On 1-5-17 I got up after sleeping 3 1/2 hours. During the last 20 minutes I had 3 Hyponeas, 2 ClearAirway, 2 Pressure & 5 PressurePulse events. I'm guessing that this wake-up was probably sleep apnea related.

On 1-7-17 I got up after only 1 1/2 hours. During that time I had 43 Pressure events & nothing else. I'm guessing that was indeed a bathroom wake-up call (I probably didn't pee before going to bed). But on the other hand, maybe something completely independent woke me up, like my cats fighting.

Once I even had a Snore event at the very second of the session's end. Surely, I couldn't wake up & turn off the CPAP within a second.