SleepyHead 0.9.8-1 Testing Builds

[Pugsy]

And if I claim that I had 0.0 in SleepyHead both yesterday and today, could it be a bug?

Sometimes the S9 machines have been known to get stuck reporting the same identical report multiple days consecutively.
While it isn't impossible to get a 0.0 AHI...it isn't all that common either but some people do get nice low less than 1.0 AHI numbers often and sometimes they do get 0.0.

For the "stuck" 0.0 AHI...all we usually have to do is unplug the machine from the electric source and then plug it back in. This forces the machine's firmware to "reboot" not unlike when we reboot a computer when it gets some cobwebs and doesn't want to play nice. I don't do this unless I see 0.0 3 nights in a row...and while I normally have less than 1.0 AHI and I do see 0.0 fairly often...it would be unusual for me to go 3 nights in a row with 0.0 AHI. I almost always have some little something of an event even if it is just one hyponea.

Since it hurts nothing to do the reboot thing..it is just something that I do when I might be suspecting the machine is stuck on 0.0.
Not long ago I went a week between downloads and the last 4 nights showed 0.0 AHI...I am good but not usually that good so I did the unplug thing and the next night the AHI was 0.56 I think it was. I can't prove the machine was stuck but I suspect it was for those 4 nights.

[tan]

And if I claim that I had 0.0 in SleepyHead both yesterday and today, could it be a bug?

Sometimes the S9 machines have been known to get stuck reporting the same identical report multiple days consecutively..

My S9 shows non-0.0 AHI (still below 1), but it is SleepyHead that got stuck

[palerider]

Detected an unusually high AHI today on my Autoset S9: 1.4. Imported data into SleepyHead: it showed 0.0. Anyone else reported the same?

Sometimes there's a difference in which "day" you're looking at, and when the "day" starts. See if there's a 24 hour time period in your SH data that matches the "1.4" AHI.

And if I claim that I had 0.0 in SleepyHead both yesterday and today, could it be a bug?

you can upload your sd card to dropbox, if you want, and send the link to someone who can take a look at the data, for a second opinion.

[Pugsy]

My S9 shows non-0.0 AHI (still below 1), but it is SleepyHead that got stuck

Sorry...I thought you meant both the S9 and SleepyHead showing 0.0 still.
If SleepyHead isn't jiving with the LCD screen then I would be using ResScan to verify and go from there.
I see PR has offered to compare... I assume he will also use ResScan to verify and confirm as much as he hates ResScan...it needs to be done. SleepyHead and ResScan and the LCD screen should all come in real close to each other with only maybe minor round up or down in tenths as a difference.

[tan]

Detected an unusually high AHI today on my Autoset S9: 1.4. Imported data into SleepyHead: it showed 0.0. Anyone else reported the same?

Sometimes there's a difference in which "day" you're looking at, and when the "day" starts. See if there's a 24 hour time period in your SH data that matches the "1.4" AHI.

And if I claim that I had 0.0 in SleepyHead both yesterday and today, could it be a bug?

you can upload your sd card to dropbox, if you want, and send the link to someone who can take a look at the data, for a second opinion.

I swear I saw above AHI > 1 today and confirm that both the yesterday and today data got imported with "0.0". Should I zip the contents of the whole card or just part of it?

Or I can download ResScan and see if there are any differences and we will it from there? If so, any links to the software

[palerider]

I swear I saw above AHI > 1 today and confirm that both the yesterday and today data got imported with "0.0". Should I zip the contents of the whole card or just part of it?

whole card.

[tan]

Yes, ResMed software seems to show the correct number: 1.4 AHI, whereas SH - 0.0.

Alternatively, it could be my Autoset S9 and its software showing an incorrect AHI for I don't see any events on ResMed charts.

[palerider]

Yes, ResMed software seems to show the correct number: 1.4 AHI, whereas SH - 0.0.

Alternatively, it could be my Autoset S9 and its software showing an incorrect AHI for I don't see any events on ResMed charts.

yeah... i loaded up your data in SH, and resscan, it looks like sleepyhead is right.

I didn't see any apneas in the SH data that were long enough to be flagged, in fact, nothing was flagged.

looking at it in resscan, IT doesn't flag anything either, but it says you had 1.4 (and the day before, 1.9 ai of 'unknown' but, no flags on the detailed graphs... so, who's ass is it pulling those numbers out of?

also, you aren't using your machine enough. 3-4 hours a night *waggles finger*

oh, and ps, what a pain in the arse, I had to copy his data onto a sd card and plug that in before resscan would deign to look at it! *slap*

[tan]

Yes, ResMed software seems to show the correct number: 1.4 AHI, whereas SH - 0.0.

Alternatively, it could be my Autoset S9 and its software showing an incorrect AHI for I don't see any events on ResMed charts.

yeah... i loaded up your data in SH, and resscan, it looks like sleepyhead is right.

I didn't see any apneas in the SH data that were long enough to be flagged, in fact, nothing was flagged.

looking at it in resscan, IT doesn't flag anything either, but it says you had 1.4 (and the day before, 1.9 ai of 'unknown' but, no flags on the detailed graphs... so, who's ass is it pulling those numbers out of?

scratching my head also... maybe some boundary condition?

also, you aren't using your machine enough. 3-4 hours a night *waggles finger*

I used to use it more a few weeks ago. But then I started experimenting with masks, auxiliary oral appliances and higher pressure. I discovered that P10 is a bit better option for it enforces nasal breathing and expands my flimsy nostrils better than BreatheRight but unfortunately and eventually irritates them to a point that I had to take it off - an adjustment period, I hope to say.

[palerider]

I used to use it more a few weeks ago. But then I started experimenting with masks, auxiliary oral appliances and higher pressure. I discovered that P10 is a bit better option for it enforces nasal breathing and expands my flimsy nostrils better than BreatheRight but unfortunately and eventually irritates them to a point that I had to take it off - an adjustment period, I hope to say.

it took me a week or two to get used to the p10 after a ffm, I used the lanolin cream on my nose and after a while found that it wasn't needed anymore. now it doesn't bother me at all.

as to why resscan and the s9 are showing ahi when there's nothing in the flow data, .... I've no clue, but it looks like SH is right to call them 0 nights.

[PKeller]

I'm new to all this so a little background might be in order. Since this thread is for SleepyHead testing SW comments I'll describe what I ran into first and then explain how I got from being a tired but ignorant engineer to a very grateful (but still tired) SleepyHead fan:-)

I want to be able to import .SpO2 files into SH but every time I tried I got an error saying "no installed module can parse the file". This has forced me to download (each morning) directly into SH via USB. By itself that's not such a hassle. Add to this removing the SD card and lining up the timestamps from the S9 and PulseOx and it becomes a bit more of a chore. I'd prefer to be able to skip the manual steps and have all the S9 and CMS50 data automatically saved on my PC so I can start my day without worrying about remembering to download before I start recording that night before I go to bed. This means importing both the S9 and .SpO2 files into SH.

After some digging around I was able to "git" the 0.9.8-1 source, QT, MinGW and some other stuff set up so I could build SH in my PC and run the QT debugger. What I found was an error in the code that parses the full path to the .SpO2 file to see if it's one that SH recognizes. Basically, it looks for the first period (".") in the path and takes everything after that as the file extension. Since the default download folder for SpO2 Assistant has a folder called "SpO2 Assistant V2.7.0" in the path, SH was taking everything after "SpO2 Assistant V2." as the file type. Since every "extension" now started with a "7.0\" no match with supported file types was found and SH errored out. I thought about just relocating the default download location so that there would be no "." before the end of the pathname, but felt it was better to figure out the problem so I didn't run into it again by accident.

The fix seems to be simple: look BACKWARDS from the end of the pathname for the first "." and compare that with the supported file types. Changing line 154 of cms50f37_loader.cpp from "QString ext = path.section(".",1);" to "QString ext = path.section(".",-1);" seems to do the trick. I submitted a bug report in SourceForge so hopefully this will make it into the next testing build.

Since I'm brand new to this forum and Apnea stuff in general I thought I'd introduce myself and explain how I came to be messing with SH in the first place:

In October, after several years of increasing daytime fatigue, I was diagnosed with severe central apnea (AHI>70, 95% of them centrals). I just started treatment a few weeks ago with a ResMed S9 and a generous amount of O2 bleed. Now the question is whether I stick with that or need to go to ASV. That means an additional in-lab study and taking a close look at the S9 detailed logs to see how well the preliminary treatment is working. Since I travel a lot (usually to sea level locations), my office is in the city (elev 6700 ft) and my home is in the mountains (9600 ft) figuring out the right treatment is complicated. Typically I'll split my time each week between the office and my home (I can work at home easily) and am out on travel every few weeks. My first 4 weeks of treatment have been typical with my time split equally between home, office and on travel. I travel a lot internationally so sleeping on the plane is normal for me.

I bought a CMS50FW and have been using it every night. Each morning I pull the S9 SD card and copy it to my PC, import the data into SH and then download the PulseOx recorded data and make sure the timestamps are all lined up. So far, I've been really pleased with the CMS50FW. I've done several experiments running it while I'm on airplanes to see how I respond to variations in cabin pressure. I even wore it today while skiing and it held up well inside my mittens. The results are really interesting, but that's another thread, I guess. The Bluetooth connection seems to drop often but that might not be the fault of the CMS50F.

I'm working on automating the downloads using a Toshiba Flash Air SD card and the Bluetooth connection on the CMS50FW. I'm hoping to be able to just get up in the morning, plug the CMS50 in to recharge, and know that all the data has been captured. Eventually I'll want to automatically fire up SH and do the imports and timestamp alignment by finding events in the SpO2 and pulse rate waveforms that correspond to events/waveforms in the S9 logs. From looking at the data so far, at least in my own case, I think this can be done with some straightforward signal analysis using SciLab or Python. If I can skip the manual downloads and import automatically from files already downloaded to my PC via Bluetooth and WiFi I'll be way ahead.

[archangle]

For tan's phantom zero problem:

Is Ignore short sessions and combine close sessions set off in SH?

Send me the link to the SD card image if you like.

[tan]

For tan's phantom zero problem:

Is Ignore short sessions and combine close sessions set off in SH?

Send me the link to the SD card image if you like.

PMed the link. "Ignore short sessions" and "combine" are actually on

[archangle]

For tan's phantom zero problem:

Is Ignore short sessions and combine close sessions set off in SH?

Send me the link to the SD card image if you like.

PMed the link. "Ignore short sessions" and "combine" are actually on

My suggestion is to turn them off and see what happens.

[tan]

For tan's phantom zero problem:

Is Ignore short sessions and combine close sessions set off in SH?

Send me the link to the SD card image if you like.

PMed the link. "Ignore short sessions" and "combine" are actually on

My suggestion is to turn them off and see what happens.

Turned them off -- nothing changed.