ozij wrote: ↑Tue Jan 16, 2024 7:09 am
Google enough and you'll find the precise time periods Resmed's algorithms use in their calculations.
As for this:
Ralph437 wrote: ↑Tue Jan 16, 2024 12:08 am
Are you in agreement that the "cough" (apparently represented by the Inspiration/exhalation at 23:16:52) is real and not an artifact?
You say she coughed. The flow shows a deep inhalation and a deep exhalation, and then some lower flow breaths. If you want to know how the algorithm's calculation are affected by that change in flow, you have to find the details of ResMed's algorithms and make the calculations. Can't send you links - must have lost them.
If what I sent you was from 2015, it was not what I meant to send.
Try this from 2022 -
https://link.springer.com/chapter/10.10 ... 06413-5_18 I remember finding version not behind a a paywall, though this one is.
Thanks,
Regarding my observation of the cough. I've been working hard to eliminate mask leaks (she's using a FFM) so we've been trying different masks and "fittings" etc to eliminate leaks. In doing so, I've been checking on her while she is asleep about every 15 minutes during her first 2 hours of sleep. It was while I was doing a check that she had one cough, so I noted the time. I had previously synchronized my watch with her Oscar record. At the time of the cough my watch read 23:16 then with a matter of seconds read 23:17. That is why I'm confident that her cough was reflected in the Oscar record at 23:16:52.
I'm reluctant to pay the 30.00 to access the article from Springer, particularly because I don't know if it will be more explicit than the description of Hypopnea detection given in the 2015 publication. Which BTW states that Resmed's H detection is: "12s RMS scaled average 25-50% for 10s with at least one obstructed breath" Based on what others (rubicon) have said, the one obstructed breath criteria is satisfied by any flow limitation index above 0.0
So, I'm planning on seeking out the authors of the article you cited from 2022 and see if some kind soul will send me a PDF of article (but I actually don't hold out a great deal of hope because they are likely under contract to not share their work unless Springer gets $$$)
Out of curiosity, I'm going to also try to integrate several of the inspiration breath curves to see what the volumes are.
(I suspect by this point you can reasonable guess that I'm a engineer ! )
