(02-13-2016, 08:26 AM)wolson Wrote:(02-12-2016, 06:29 PM)Sleeprider Wrote: Read page 7 of 28, "Detection accuracy of obstructed airway apnea apneas (OA) and clear airway apneas (CA) by proprietary algorithm in positive airway pressure (PAP) devices. https://www.usa.philips.com/b-dam/b2bhc/...0FINAL.pdf
The pressure pulse used to detect the type of apnea is visible as a flow fluctuation in CA.
Thank you: that answers the question when I do see that pulse and rebound. I had not seen that manual and probably neither had my pulmonologist. Thus that small pulse is machine generated, not subject generated. I also tested it last night but I haven't yet looked at the data but I clearly heard a quick change in the pitch of the machine at the 10 second mark.
However, it does not answer the question of CA scoring where there is no pulse and reboubd. This can not be the only part of the algorithm that detects and scores CA's.
There are some criteria that probably needs more exploration including the 40% and 80% reductions in flow from the baseline level.
After looking at the test data of the test that I ran last night, it is not as clear as I thought it was. The pulses are very small, so small that they do not show up on the flow results until one zooms on them using the Y axis scaling. Thus, I can not conclude that the pulse and rebounds shown in the original charts are in fact the pressure pulses of the machine but I think they certainly coincide with them.
I did not realize that machine actually marked the pressure pulses as events and can be show on the charts. Note in the second test, the machine did not record a pressure pulse although it appears that one may have occurred! However there was no rebound associated with it.