SleepyHead: Updated Version 0.7-4229 Alpha

Hi guys, thanks for the wonderful work.

Pugsy, introduced me to SleepyHead, after countless number of failed
attempts to install Encore on my Japanese 32-bit Windows XP SP3 PC.

I was able to successfully start SleepyHead 0.8.0 but I have one minor
problem at the moment. Some characters are not displayed correctly. See the
red circled/lined portion of the following image.


Uploaded with ImageShack.us


Uploaded with ImageShack.us

This maybe of font issue (I have July in Japanese on the calendar, see green circle on the overview image) , what can I do to fix this problem?

sayg wrote:Hi guys, thanks for the wonderful work.

Pugsy, introduced me to SleepyHead, after countless number of failed
attempts to install Encore on my Japanese 32-bit Windows XP SP3 PC.

I was able to successfully start SleepyHead 0.8.0 but I have one minor
problem at the moment. Some characters are not displayed correctly. See the
red circled/lined portion of the following image.


Uploaded with ImageShack.us


Uploaded with ImageShack.us

This maybe of font issue (I have July in Japanese on the calendar, see green circle on the overview image) , what can I do to fix this problem?

I have made no attempt to internationalize SleepyHead yet, as I want to get the functionality working first.. I've have however been paying attention to the translation stubs, so when I do get around to figuring out how to get it to work, it will be ready for willing translators..

At the moment it is supposed to use it's integrated FreeSans font for everything. Which is not working very well in this case. The problem with fonts in an interactive graphing application like this one, is it is incredibly difficult to get consistency regarding where text is placed.

It should still be rendering correctly in english, there will be a locale specific thing interfering with the way it works.. I'll look into it and see what I can find out.
This could also be a Japanese version specific WinXP problem, or alternatively an issue with your video card.

Jedimark, thanks for a quick response.

jedimark wrote: I have made no attempt to internationalize SleepyHead yet, as I want to get the functionality working first.. I've have however been paying attention to the translation stubs, so when I do get around to figuring out how to get it to work, it will be ready for willing translators..

I agree with your strategy of working on the functionality first. With that said, I think it's great that you are thinking multilingual support. Here in Japan, if we go through the regular channels, only doctors are only allowed to see data. I ask them to print out a report but they refuse. Software are not sold to patients so we(patients) have basically no options. So I think it will help the international (or at least the Japanese) community tremendously. When you get the multilingual support going, I can do the Japanese part so please let me know. But first, I'll have to understand the data,,, I don't understand most of the data shown in SleepyHead. Does anyone have any resources I can reference?

jedimark wrote:At the moment it is supposed to use it's integrated FreeSans font for everything. Which is not working very well in this case. The problem with fonts in an interactive graphing application like this one, is it is incredibly difficult to get consistency regarding where text is placed.

It should still be rendering correctly in english, there will be a locale specific thing interfering with the way it works.. I'll look into it and see what I can find out.
This could also be a Japanese version specific WinXP problem, or alternatively an issue with your video card.

Thanks for looking into it. When you say "integrated FreeSans", do you meanFreeSans fonts are embedded into the Sleepyhead binary? I checked my system fonts but there were no freesans fonts. It might very well be a locale problem, since I have July showing in Japanese word rather than English word "July".

sayg wrote:Jedimark, thanks for a quick response.
I agree with your strategy of working on the functionality first. With that said, I think it's great that you are thinking multilingual support. Here in Japan, if we go through the regular channels, only doctors are only allowed to see data. I ask them to print out a report but they refuse. Software are not sold to patients so we(patients) have basically no options. So I think it will help the international (or at least the Japanese) community tremendously. When you get the multilingual support going, I can do the Japanese part so please let me know. But first, I'll have to understand the data,,, I don't understand most of the data shown in SleepyHead. Does anyone have any resources I can reference?

Thanks for looking into it. When you say "integrated FreeSans", do you meanFreeSans fonts are embedded into the Sleepyhead binary? I checked my system fonts but there were no freesans fonts. It might very well be a locale problem, since I have July showing in Japanese word rather than English word "July".

Doctors over in Japan need to lighten up a bit.. It's your data, and you are entitled to hold copyright over it. You signed no release waivering this right.

Once I figure out how the translation stuff works, I'll post a call for translators on this forum.

I'm kinda glad you brought this up anyway, as I need to start looking into this.. English speaking people aren't the only ones who deserve better OSA treatment.

Yes, FreeSans font is part of the binary.. It's meant as more of a temporary thing while I figure out how to make text draw code consistently. It gets loaded when the app starts up, so is supposed to work regarless of locale. :/

I'm starting to suspect this is a windows XP specific problem. Do you have access to another PC you could test it on?

sayg wrote: But first, I'll have to understand the data,,, I don't understand most of the data shown in SleepyHead. Does anyone have any resources I can reference?

I copied the entire glossary out of Encore Pro and put it in a post some where. I could copy it again if it would help you understand the terms. Unfortunately there isn't anything that says so and so needs work. We just have to figure all this out on our own. Are you needing help with terms and definitions or understanding the values shown?

Edit.. Regarding the values shown on the single night data. Image is a bit small and I can't see overall numbers but in general those are quite good. Target AHI less than 5, you meet that easily. I suspect the bulk of the snores were actually mask leak noise as you can see they correlate as to times. Also from the pattern to the bulk of the snores.
Since we can't see leak number actual values on the graphs and the other numbers are too small to read (above the pie) I can't tell exactly how large the leak is. All in all a quite decent night as far as numbers go.

Question -- the flags on my Sleepyhead seem a little off. See pic below. The flag is after the OA, almost a minute after it occurs. I imported the data into Encore, and the flags are right in there. Unless I'm totally reading the waveform wrong...



Hmm...the image isn't showing up. Try the URL:

http://twitpic.com/5sy4n5/full

nudave wrote:Question -- the flags on my Sleepyhead seem a little off. See pic below. The flag is after the OA, almost a minute after it occurs. I imported the data into Encore, and the flags are right in there. Unless I'm totally reading the waveform wrong...



Hmm...the image isn't showing up. Try the URL:

http://twitpic.com/5sy4n5/full

I've got a few events where this is where the machine put the events.. I have no idea why, but it's flagged this way.

In the few cases where I have observed this, Encore Pro's waveform overlay did the same thing.. :/

Edit: Can you get me a screenshot of Encores doing this? Also, perhaps could I have a copy of the .001/.002 & .005 session files belonging to that day.? You may have just discovered something we don't know yet..

I've got a few events where this is where the machine put the events.. I have no idea why, but it's flagged this way.

In the few cases where I have observed this, Encore Pro's waveform overlay did the same thing.. :/

Edit: Can you get me a screenshot of Encores doing this? Also, perhaps could I have a copy of the .001/.002 & .005 session files belonging to that day.? You may have just discovered something we don't know yet..[/quote]


Here's the Encore SS of the same event:

http://twitpic.com/5sypdm/full

Any idea on where I can upload the raw data files for free?

EDIT: Or PM an email address so I can attach them..

It says 2:38 there.. each bar in waveform viewer represents 1 minute, 6 minutes per line.. You may be looking at the wrong one..

the event in question should be nearer to 3:41-3:42

Edit: Unless daylight saving issue is coming into play.
Edit Edit: It very much looks like this is the case..

Make sure you've got the right day in encore.. SleepyHeads session start date (not necessarily the calendar for night owls) should match the Encore reports date.
(It pretty much works the same way as Encore now, it's just been a while since I played..)

jedimark wrote:It says 2:38 there.. each bar in waveform viewer represents 1 minute, 6 minutes per line.. You may be looking at the wrong one..

the event in question should be nearer to 3:41-3:42

Edit: Unless daylight saving issue is coming into play.


Make sure you've got the right day in encore.. SleepyHeads session start date (not necessarily the calendar for night owls) should match the Encore reports date.
(It pretty much works the same way as Encore now, it's just been a while since I played..)

Yes, it is the right one. I only had one OA last night. For some reason, I can't get Encore to provide the correct time, with the DST settings.

nudave wrote: Yes, it is the right one. I only had one OA last night. For some reason, I can't get Encore to provide the correct time, with the DST settings.

It won't. Bug in Pro software I guess. It will be one hour off till fall back to regular time, at that time it will correct itself. Changing to DST off makes it 2 hours off. I have tried all I know to correct this issue and can't get it done short of really messing with clock on computer and or time zones in Pro.

Yay.. It was not an error in the event codes.. (thank goodness, because that would have sucked..)

The timing error was in the duration of the first waveform chunk. The starting timestamp+chunk duration did not equal the second waveform chunks starting timestamp. It only appears to happen on the first waveform chunk entry.. So all I had to do was sync the starting time when this happens. It was out 60 seconds neat, which kinda indicates this was designed that way for some reason..

This is fixed now in the source code.. Check the other 0.8 thread for an update in the next day or two.. I've got some ASV stuff to unbreak before I do that though..

The 0.8 thread is a better place to continue stuff on the Qt branch.. This 0.7 is old and dead.. It's nearly a different program there was that many changes :}

Jedimark, apologizes for the slow response.

jedimark wrote:I'm starting to suspect this is a windows XP specific problem. Do you have access to another PC you could test it on?

Unfortunately, I do not have access to PCs other than Win32 XP SP3 right at this time. My father has a Windows 7 machine (I think it's 64-bit), so I test it there when I get a chance.

Pugsy, so sorry for the late response.

Pugsy wrote:I copied the entire glossary out of Encore Pro and put it in a post some where. I could copy it again if it would help you understand the terms. Unfortunately there isn't anything that says so and so needs work. We just have to figure all this out on our own. Are you needing help with terms and definitions or understanding the values shown?

I do need help but I think I found your glossary post, http://bit.ly/odcvsK .
I will keep it for reference. Thanks.

Pugsy wrote:Edit.. Regarding the values shown on the single night data. Image is a bit small and I can't see overall numbers but in general those are quite good. Target AHI less than 5, you meet that easily. I suspect the bulk of the snores were actually mask leak noise as you can see they correlate as to times. Also from the pattern to the bulk of the snores.
Since we can't see leak number actual values on the graphs and the other numbers are too small to read (above the pie) I can't tell exactly how large the leak is. All in all a quite decent night as far as numbers go.

Thanks for the analysis. FYI, the numbers for that single night data is as follows:

AHI 5.75 RERA 3.20
Hypopnea 4.48 FlowLimit 0.00
Obstructive 0.64 Vsnore 14.71
ClearAirway 0.64 PB/CSR 0.00%

90% Pressure 9.40
Min Avg Max
Pressure 6.00 9.54 10.00
Leak 19.00 43.70 164.00
Snore 0.00 4.15 21.00

Date Start End
2011/07/13 02:46 04:20

My doctor tells me that I am always taking off the mask 1-2 hrs after falling asleep. I still hate waking up, I never feel good when waking up, I am so sleepy in the daytime. And I am within Target AHI, I can't imagine what it's like for ppl over Target AHI.

Also, I monitor my sleep quality using actigraphy device fitbit and it tells me that for that night, I had a sleep efficiency of 76 %. I usually get 40%-60% so I guess it was one of the better nights.

Fitbit details:

You went to bed at 02:38
Time to fall asleep 13min
Times awakened 34
You were in bed for 4hrs 44min
Actual sleep time 3hrs 25min

Cheers.

sayg wrote:My doctor tells me that I am always taking off the mask 1-2 hrs after falling asleep. I still hate waking up, I never feel good when waking up, I am so sleepy in the daytime. And I am within Target AHI, I can't imagine what it's like for ppl over Target AHI.

sayg wrote:Times awakened 34
You were in bed for 4hrs 44min
Actual sleep time 3hrs 25min

The above points to very frequent awakenings along with obviously short hours of sleep. The best AHI numbers in the world will have a pretty steep uphill battle to feel good as long as these factors are occurring.

Frequent awakenings mess with sleep architecture. Prevents the body from going through the normal cycles that are needed for the body to recoup. Add in short hours of sleep and you have a recipe that pretty much sets a person up for feeling crappy. I have been down this road myself.

You need to try to figure out what is causing the awakenings and try to limit them.
Pain? Meds? Bed comfort? Anxiety? Mask issues? Any number of possible causes.
Gotta work on staying asleep and hope that let's you sleep longer.