Arik - Therapy Thread

[sopherFellow]

If I was tasked with optimizing AHI detection for a CPAP firm, I could definitely envision deriving a sophisticated algorithm, for which the definition of a valid or significant hypopnea would be much more complicated than (x% flow reduction for y seconds).   I would gather pulseOx data and Quality of Sleep surveys and I don't know what other data and try to quantify severity of event in terms of O2 saturation and PQS (Perceived Quality of Sleep) etc., based on factors such as super detailed analysis of the event (more sophisticated than x% reduction for y secs), perhaps data immediately preceding and immediately following the event.   If any correlation could be demonstrated to severity of the hypopnea with these factors, I would try to use this to establish an predictive algorithm that could be used based on CPAP data alone (without explicit oximetry or PQS data).  I would recommend to my company that this more detailed approach be adopted for reported events in order to help the lay user focus on events that are more important.   I would also closely guard the details or even the existence of such an algorithm, as it would be intellectual property which could be a significant competitive advantage.

And we haven't even discussed mitigation, like pulsating pressure during the event!

I'm not saying that this is what ResMed is doing, but I don't know that it isn't.  I'm new to this forum but 2 things I have heard a lot here is that ResMed is different from the other machines and the overwhelming consensus is that they are better.  My interpretation is that people are saying that they feel better the next day after using an AirSense S10 - I do and isn't that the real object of the exercise?  If any part of my conjecture is true, I'm happy not to chase after non-significant events.   Draw your own conclusions about whether we should worry about simplistic x% reduction for y secs events

[Gideon]

And what is the basis for a normal breath? 50% is the border between flow limitations and hypopneas. Most of the breaths in your closeup view appear to be either shallow or flow limited. What does your normal flow look like?

[Arik]

If I was tasked with optimizing AHI detection for a CPAP firm, I could definitely envision deriving a sophisticated algorithm, for which the definition of a valid or significant hypopnea would be much more complicated than (x% flow reduction for y seconds).   I would gather pulseOx data and Quality of Sleep surveys and I don't know what other data and try to quantify severity of event in terms of O2 saturation and PQS (Perceived Quality of Sleep) etc., based on factors such as super detailed analysis of the event (more sophisticated than x% reduction for y secs), perhaps data immediately preceding and immediately following the event.   If any correlation could be demonstrated to severity of the hypopnea with these factors, I would try to use this to establish an predictive algorithm that could be used based on CPAP data alone (without explicit oximetry or PQS data).  I would recommend to my company that this more detailed approach be adopted for reported events in order to help the lay user focus on events that are more important.   I would also closely guard the details or even the existence of such an algorithm, as it would be intellectual property which could be a significant competitive advantage.

And we haven't even discussed mitigation, like pulsating pressure during the event!

I'm not saying that this is what ResMed is doing, but I don't know that it isn't.  I'm new to this forum but 2 things I have heard a lot here is that ResMed is different from the other machines and the overwhelming consensus is that they are better.  My interpretation is that people are saying that they feel better the next day after using an AirSense S10 - I do and isn't that the real object of the exercise?  If any part of my conjecture is true, I'm happy not to chase after non-significant events.   Draw your own conclusions about whether we should worry about simplistic x% reduction for y secs events

Hi SopherFellow.
These are words of wisdom. However, let me tell you that I felt great after a night of full PSG test in hospital while the report claimed 144 (!) arousals which I was not aware of and 68 disaturation events. There is a direct link between high AHI and plenty of disaturation events that can be detected at home with a $200 device. I suspect no one can tell the difference in feeling after a night with 100 or 20 disaturation events. 
ResMed is by far the most comfortable and quite device I tested and I plan to keep using it.  What I currently have in mind is to optimize my therapy towards  minimizing the number of events, no matter of what type or length of time, and check periodically my saturation profile (cannot do this every night). The great feature in OSCAR of defining user's customized events will enable me doing this without the need to spend hours looking at graphs.
Thanks,
Arik

[Arik]

And what is the basis for a normal breath? 50% is the border between flow limitations and hypopneas.  Most of the breaths in your closeup view appear to be either shallow or flow limited.  What does your normal flow look like?

Hi Fred,

The attached is a typical "good" one. I noticed that even at nights with very low AHI, I still have a lot of flow limitations. Can I do something about that?

BTW, OSCAR is a GREAT application!

Thanks,
Arik

[Arik]

And what is the basis for a normal breath? 50% is the border between flow limitations and hypopneas.  Most of the breaths in your closeup view appear to be either shallow or flow limited.  What does your normal flow look like?

Hi Fred,

The attached is a typical "good" one. I noticed that even at nights with very low AHI, I still have a lot of flow limitations. Can I do something about that?

BTW, OSCAR is a GREAT application!

Thanks,
Arik

Forgot the attachment...

[sopherFellow]

Arik

I recently got the Wellue O2Ring for $170 on Amazon and now I monitor my O2 every night, in case you want to do that.

Dennis

[Arik]

Arik

I recently got the Wellue O2Ring for $170 on Amazon and now I monitor my O2 every night, in case you want to do that.

Dennis

It will be interesting to have your opinion about the Wellue. I'm using the CMS 50F plus and trust mainly relative values, rather than absolute ones. The display is very "jumpy" so it is hard to get a real time value but the written overnight report is good and you can have trends of average saturation value and number of disaturation events.
Arik

[Gideon]

You have maxed out your pressure support by having EPR=3. I would set your min pressure to 7 as that will maximize your benefit at low pressures. That means your min pressures IPAP/EPAP = 7/4

[sopherFellow]

Additional response to your comments about user insensitivity to events.

On a typical night I have zero events where my O2 drops below 90%.   Nonetheless, I feel substantially more or less rested/tired on a day following a high/low number of events reported by my AirSense.  O2 saturation is important, but it is not the whole story.  Disruptive sleep, even if it does not reduce O2, may still prove to be unsatisfactory and non-restful.

You may be particularly insensitive to events - this might be because you are used to having events and you may have such a large sleep deficit that it will take quite a while for you to see improvement.   So, I disagree that users in general are insensitive to events.   I see a key question asked in virtually every thread on here (including this one) is "How are you feeling next day?"   Once you overcome your sleep deficit acquired over a long period of time, perhaps you will find this a useful question as well.

Submitted for your consideration.

Dennis

P.S.   Early days, but so far very impressed with the O2Ring.  On the 1 occasion that I had an O2 drop, it's vibration alarm awoke me and I was able to fix the problem.

[sopherFellow]

   
Here's a screenshot of last night's results from the O2Ring.    Note the drop in O2 around 3:00 - here the vibration alarm woke me and I adjusted my collar.