Problem with SleepyHead Oximeter Data import

[Arizona-Willie]

When I import breathing data into Sleepyhead it graphs it correctly.

When I impor 02 data from my Contec p/o into Sleepyhead it graphs it many minutes too long --- up to 20 sometimes.

When i import the 02 data into SPO2 it graphs it correctly and when I take the starting time and add the running time it shows it ending at the correct time.

This used to work correctly in Sleepyhead too but started malfunctioning recently.

This is with version 1.0.0 Beta 2

Is there any way to fix this?

[Arizona-Willie]

This problem began when I set the time on my Resmed S9 APAP to match the time on my CMS 50-F wrist pulse oximeter.

Before that the time was off on the start, now it's off at the end.

[Jay Aitchsee]

What makes you think your CMS recording session is not actually longer than your CPAP session? Did you start and end both at the same time? If one session (CMS or CPAP) is longer than the other, the longest will determine the length of the display. It's up to you to determine how to synch the start times. There are a couple of ways depending on the model CMS.

Just judging by some of the spikes in the flow, plulse, and SPO2 graphs, it looks like they might be pretty close. It looks to me like you shut off the cpap about 7:00 but didn't stop the CMS recording until nearly 7:30. That's not what happened?

[palerider]

This is with version 1.0.0 Beta 2

Is there any way to fix this?

either a) hurl rocks at eastern australia until watkins gets off his ass and fixes his buggy program, or b) encourage other people to start working on the program and fix the bugs or c) if you're a c++ programmer, fix them yourself, and earn the adoration of the community

[Arizona-Willie]

I get the menu on my PO set to Record but wait to push the button until I push the button to start the APAP machine. I start both within one second because I push both buttons at the same time.

I set the time on the APAP to match the time on the PO because before the time on the APAP was off by about 20 minutes or so. And the graphs were off at the start but now they are off at the end.

But the APAP and PO both show the same time now.

I also record the PO into SPO2 and everything there works just fine. If I add the run time to the start time I get a correct end time --- just like it should be.

What makes you think your CMS recording session is not actually longer than your CPAP session? Did you start and end both at the same time? If one session (CMS or CPAP) is longer than the other, the longest will determine the length of the display. It's up to you to determine how to synch the start times. There are a couple of ways depending on the model CMS.

Just judging by some of the spikes in the flow, plulse, and SPO2 graphs, it looks like they might be pretty close. It looks to me like you shut off the cpap about 7:00 but didn't stop the CMS recording until nearly 7:30. That's not what happened?

[Jay Aitchsee]

So, your saying you pressed the CMS record off at the same time as you pressed CPAP off button? Just as you did when you turned both on at the start of the night? It sure looks like you left the CMS recording running for about 20 minutes after you shut off the CPAP. There's a slight elevation in Pulse rate and SPO2 which would be normal under those circumstances. Again, the start and stop session times of whatever's imported, CPAP, Zeo, Pulse Ox, or whatever will determine the width of all the graph windows in time.

For example, If you left the CMS recording run for an hour more than the CPAP and had synced them at the start. The flow graph window would be the same width as the pulse ox graph windows, but the flow wave graph would stop an hour before the right edge of its window while the pulse ox graphs would continue to the end. Similar to your example.

The imported data with the earliest start time opens all the windows at that time and the data with latest stop time closes all the windows at that time. Not all graphs will totally fill each of their respective windows unless it happens that all of the sessions are exactly equal in time which is unlikely.

[Arizona-Willie]

Correct I push the buttons on both the PO and the APAP machine at the same time to start them and to stop them.

Before I corrected the time on the APAP the start times did not match up but since then they do but the end times don't match up --- but the end times in the program SPO2 < DO > match up.

I take the start time say 10:30:20 and add the running time say 7:32:10 and get an end time of 6:03:00.

You have to take the start time and subtract that from midnight 12:00:00 which, in this instance, gives us 1:20:40 ( time before midnite ) and then subtract that from the total run time to get 6:03:00 which would be the stop time / wake time.

The times in SPO2 work out perfectly but they get messed up by 20 some minutes when imported into Sleepyhead.

I don't know what is wrong with Sleepyhead that causes this but since the times are correct in the other program and the time settings on the PO and the APAP match it pretty well < has > to be a bug in the program.

And that bug has been there because the start times were off before I corrected the time in the APAP.

Fixing the time in the APAP just shifted the error to the end of the recording instead of the beginning.

I have an appointment this week with the oximetry clinic and I am going to show them the results from SPO2 --- not Sleepyhead because of this bug.

[Jay Aitchsee]

Willie, I guess you have a bug and PR must think so as well, judging by his comments above.

I have not experienced it with SH ver 0.9.8-1. Apparently it has popped up with the newer version you are using.

Not that I think it is the problem, but you might check to make sure there is no Time offset under SH/File/Preferences/CPAP

Because I have to upload my CMS 50F data to the SPO2 assistant first and then fetch the file from there with SleepyHead, I assume you do too?
With my set-up, I sync the CMS50F time to the computer, then I sync the CPAP time to the CMS50F. In use, I start and stop the two close in time, but I don't worry about trying to do it simultaneously. When I import the CMS data into SleepyHead, I use the option "I want to use the time reported by my oximeter's built in clock", since the oximeter does indeed have a clock and it has been synced with the CPAP clock already.

This method will result in the data of the CPAP and CMS being synched within a few seconds, as close as the original time synching between the two. I find this to be as accurate as trying to start the two simultaneously and then using the other import option to sync. The difference will be as I've outlined in the post above - the two sources will have different start and stop times in their respective windows as they should, but the events of each will be matched in time (within a couple seconds). I have to resync every few days.

Using the alternate method, which I think you use, forces Sleepyhead to reconstruct the CMS timeline and that may be where the error is introduced. Why don't you try using the option, "I want to use the time reported by my oximeter's built in clock" and see if the error persists?

Another place to look just for info is at the bottom of the SleepyHead Left Daily Sidebar under Sessions. There, the CMS Session is listed with start, end, and duration times which should correspond to the SP02 report. My duration time here is within 1 second of the SPO2 report duration time.

BTW, if you don't like the results of the syncs with either method, you can import the file again and tweak the start time a few seconds using the manual adjustment. This results in a second session being imported and displayed on top of the first (ver 9.8-1). In the Left sidebar under Sessions, either can be turned off or on by clicking on the corresponding flag. If you want to be rid of one completely, you can go to SleepyHead Data testing/profile/name/CMS50xxxx/ Events and Summaries and delete the appropriate files.

[rkl122]

................................
I don't know what is wrong with Sleepyhead that causes this but since the times are correct in the other program and the time settings on the PO and the APAP match it pretty well < has > to be a bug in the program.
....................

Hi. SH oximetry imports, direct and indirect (ie. via the SpO2 utility), are known to have issues. The exact nature of the issue appears dependent on the versions of both SH and the oximeter. Upon direct import, my CMS-50i gives time line syncing issues similar to that reported here, and SH (1.0.0-beta 2) will not import the SpO2 file. Comparable problems are reported in the bug tracker: https://bugs.jedimark.net/login_page.php

There is also this link: http://www.apneaboard.com/wiki/index.ph ... SleepyHead, which I think contains useful information, and which contains this statement: "Newer pulse oximeters (such as the CMS-50i and the CMS-50iW) do not use the same SpO2 software and do not save in the format required by SleepyHead and are not importable." (I have the version of SpO2 that works with the CMS-50i.)

Don't know where the CMS-50F fits on the spectrum of problems (seems to work ok for Jay using prior version of SH), but I'd guess your problem is related to the others.

Incidentally, by reviewing the dates of status changes in the bugtracker, one can infer when JediMark is active - or at least attending to bug reports. At present, looks like there's been no activity since mid April.

-Ron

[Arizona-Willie]

No I use the import directly from device option.

I hookup my CMS 50-F to a USB port and read info directly from the device.

Willie, I guess you have a bug and PR must think so as well, judging by his comments above.

I have not experienced it with SH ver 0.9.8-1. Apparently it has popped up with the newer version you are using.

Not that I think it is the problem, but you might check to make sure there is no Time offset under SH/File/Preferences/CPAP

Because I have to upload my CMS 50F data to the SPO2 assistant first and then fetch the file from there with SleepyHead, I assume you do too?
With my set-up, I sync the CMS50F time to the computer, then I sync the CPAP time to the CMS50F. In use, I start and stop the two close in time, but I don't worry about trying to do it simultaneously. When I import the CMS data into SleepyHead, I use the option "I want to use the time reported by my oximeter's built in clock", since the oximeter does indeed have a clock and it has been synced with the CPAP clock already.

This method will result in the data of the CPAP and CMS being synched within a few seconds, as close as the original time synching between the two. I find this to be as accurate as trying to start the two simultaneously and then using the other import option to sync. The difference will be as I've outlined in the post above - the two sources will have different start and stop times in their respective windows as they should, but the events of each will be matched in time (within a couple seconds). I have to resync every few days.

Using the alternate method, which I think you use, forces Sleepyhead to reconstruct the CMS timeline and that may be where the error is introduced. Why don't you try using the option, "I want to use the time reported by my oximeter's built in clock" and see if the error persists?

Another place to look just for info is at the bottom of the SleepyHead Left Daily Sidebar under Sessions. There, the CMS Session is listed with start, end, and duration times which should correspond to the SP02 report. My duration time here is within 1 second of the SPO2 report duration time.

BTW, if you don't like the results of the syncs with either method, you can import the file again and tweak the start time a few seconds using the manual adjustment. This results in a second session being imported and displayed on top of the first (ver 9.8-1). In the Left sidebar under Sessions, either can be turned off or on by clicking on the corresponding flag. If you want to be rid of one completely, you can go to SleepyHead Data testing/profile/name/CMS50xxxx/ Events and Summaries and delete the appropriate files.

[Arizona-Willie]

UPDATE UPDATE UPDATE

This morning I just imported the O2 file from SPO2 instead of importing directly from the device ( CMS 50-F) AND IT WORKED PERFECTLY!!!!

Importing from the device gives messed up graphing, as I've been posting, but importing from the SPO2 file imports just fine.

Looks like I have to change my order of doing things until / if ever they fix this bug.

I imported the Oximetry Report ... not the entire file and it worked.

[Jay Aitchsee]

I hookup my CMS 50-F to a USB port and read info directly from the device.

my CMS-50i gives time line syncing issues similar to that reported here

Interesting, My CMS50F, with a blue face., firmware ver 3.8, has no provision for direct upload to SH. Uploading is not an option in the menu. When connected directly via USB, SH doesn't find it and says something like "waiting for upload."

The file structure of this version of the CMS5OF has changed I think, as least from that of the CMS50E. The software I used to read the 50E would not read the 50F.

I went to Cooper Medical and chose a software that looked reasonable (Windows 7, 64bit) for my Windows 10 application, which was the SpO2 Assistant Version 2.1. and it works well. With the CMS50F connected via USB and turned on, the software will find it and provides options to upload and then subsequently review and store a record. Once the record is uploaded, I point SH to it via the "Import from a data file..." option. And then get sync options previously mentioned, and, also as previously mentioned (or implied), I am able to sync this with the CPAP data down to the second, if I wanted to spend the time to do so.

I mention all this for the possibility that it might offer a clue to an alternate method of importing Oximeter data to SH, such as through different SPO2 software.

Well, I see by the below my clue is a little late, but I spent some time composing it, so I'll post it anyway

UPDATE UPDATE UPDATE

This morning I just imported the O2 file from SPO2 instead of importing directly from the device ( CMS 50-F) AND IT WORKED PERFECTLY!!!!

Importing from the device gives messed up graphing, as I've been posting, but importing from the SPO2 file imports just fine.

Looks like I have to change my order of doing things until / if ever they fix this bug.

I imported the Oximetry Report ... not the entire file and it worked.

[palerider]

UPDATE UPDATE UPDATE

This morning I just imported the O2 file from SPO2 instead of importing directly from the device ( CMS 50-F) AND IT WORKED PERFECTLY!!!!

Importing from the device gives messed up graphing, as I've been posting, but importing from the SPO2 file imports just fine.

Looks like I have to change my order of doing things until / if ever they fix this bug.

I imported the Oximetry Report ... not the entire file and it worked.

you could also go back to 098-1, I don't remember much of any fixes in beta2 that affect the s9 series.