How does PAP differentiate between central and obstructive?

[abeck99]

The subject pretty much lines it out - I can't wrap my head around how a PAP machine, which is basing everything simply on pressure, can tell the difference between an "open airway" or closed. It seems that you need outside data for that. Is there some proprietary machine learning magic sauce developed by PAP manufacturers to differentiate with a good degree of certainty?
I'm asking because the flow data for central vs obstructive events that I have seen on sleepyhead can appear visually very similar on the flow rate.
I have a suspicion this is also related to the fact that sleepyhead doesn't call them "central" apneas.

Related question about sleepyhead - does sleepyhead itself analyze the flow or are the recorded events coming straight from the PAP machine? What publicly available information is available on analyzing flow data (I couldn't find any after a brief search on google and thought someone here might know better).

Thanks!

[ChicagoGranny]

How does CPAP differentiate between central and obstructive?

http://www.resmed.com/int/assets/docume ... -paper.pdf

[Wulfman...]

The subject pretty much lines it out - I can't wrap my head around how a PAP machine, which is basing everything simply on pressure, can tell the difference between an "open airway" or closed. It seems that you need outside data for that. Is there some proprietary machine learning magic sauce developed by PAP manufacturers to differentiate with a good degree of certainty?
I'm asking because the flow data for central vs obstructive events that I have seen on sleepyhead can appear visually very similar on the flow rate.
I have a suspicion this is also related to the fact that sleepyhead doesn't call them "central" apneas.

Related question about sleepyhead - does sleepyhead itself analyze the flow or are the recorded events coming straight from the PAP machine? What publicly available information is available on analyzing flow data (I couldn't find any after a brief search on google and thought someone here might know better).

Thanks!

The machine analyzes and stores the raw data. Sleepyhead (and ResScan) convert it into formats that we can read.


Den

.

[palerider]

I have a suspicion this is also related to the fact that sleepyhead doesn't call them "central" apneas.

yeah, you'd be wrong about that.

sleepyhead was originally written for philips respironics machines, and they use their own terminology for things, like 'clear airway', and mark just mimicked that when he wrote sleepyhead.

resmed DOES call the same situation centrals.

does sleepyhead itself analyze the flow or are the recorded events coming straight from the PAP machine? What publicly available information is available on analyzing flow data (I couldn't find any after a brief search on google and thought someone here might know better).

sleepyhead reports the data that comes from the machine, the machines insert flags into the data for the events that the machines detect.

SH reports some of the data directly, and it creates some of the data displays based on the flow, pressure and event data.

if you did around, you can find quite a bit of information by looking at the patents filed by the various machine makers, look on the bottom of your unit for patent numbers to get you started

also, youtube has nice resmed videos demonstrating how the machines determine and respond to various things.

happy googlin.

[abeck99]

Thanks everyone! This gives me a lot to read into.

That paper is very interesting, FOT is a pretty cool idea.
It's not EXACTLY determining central vs obstructive, actually closed vs open, but the appropriate therapeutic response for a closed central is the same as an obstructive so it's a moot point. From my sleep lab report, the techs marked the closed centrals as "mixed" so depending on semantics it could be considered an OA.
"By definition, all obstructive apneas are associated with a closed upper airway, but some central apneas can be associated with a closed airway if the pressure is below the airway closing pressure. If the airway is closed during a central apnea the appropriate algorithmic response is to increase the therapeutic pressure"