Hello Guest, Welcome to Apnea Board !
As a guest, you are limited to certain areas of the board and there are some features you can't use.
To post a message, you must create a free account using a valid email address.

or Create an Account


New Posts   Today's Posts

flow limitations expressed as an index?
#11
RE: flow limitations expressed as an index?
What are you hoping to achieve with all this new-found data?
Crimson Nape
Apnea Board Moderator
www.ApneaBoard.com
___________________________________
Useful Links -or- When All Else Fails:
The Guide to Understanding OSCAR
OSCAR Chart Organization
Attaching Images and Files on Apnea Board
Apnea Helpful Tips

INFORMATION ON APNEA BOARD FORUMS OR ON APNEABOARD.COM SHOULD NOT BE CONSIDERED AS MEDICAL ADVICE. ALWAYS SEEK THE ADVICE OF A PHYSICIAN BEFORE SEEKING TREATMENT FOR MEDICAL CONDITIONS, INCLUDING SLEEP APNEA. INFORMATION POSTED ON THE APNEA BOARD WEB SITE AND FORUMS ARE PERSONAL OPINION ONLY AND NOT NECESSARILY A STATEMENT OF FACT.
Post Reply Post Reply






Donate to Apnea Board  
#12
RE: flow limitations expressed as an index?
I've moved this thread into the Software Support Forum, since (I think) it likely fits better in that forum.

Thanks.   Coffee
SuperSleeper
Apnea Board Administrator
www.ApneaBoard.com


INFORMATION ON APNEA BOARD FORUMS OR ON APNEABOARD.COM SHOULD NOT BE CONSIDERED AS MEDICAL ADVICE. ALWAYS SEEK THE ADVICE OF A PHYSICIAN BEFORE SEEKING TREATMENT FOR MEDICAL CONDITIONS, INCLUDING SLEEP APNEA. INFORMATION POSTED ON THE APNEA BOARD WEB SITE AND FORUMS ARE PERSONAL OPINION ONLY AND NOT NECESSARILY A STATEMENT OF FACT.




#apneaboard
Post Reply Post Reply
#13
RE: flow limitations expressed as an index?
CN, I don't think it's new-found because it appears the data for a simple expression of 'density', whether frequency, duration or both, already exists. I'm probably repeating myself below because I think I've already addressed your question to the extent I'm able.

there's quite a bit of interest around the forum in trying to understand flow limitations, how they're defined and identified, how they might be characterized/classified, what characteristics and thresholds are significant, and generally how to better treat them.

a simple start on a much more complicated topic would be to know how much of the time my machine says my breathing is flow limited. the current stat reports min med 95%tile and max level of severity with (apparently) no consideration of time/duration.

I doubt treatment would be any different but a simple measure of 'density' (whether frequency, duration or both; and distinguished from 'intensity' or 'severity') would be another useful measure of how seriously flow limitations may be affecting us.

currently we treat to attempt to reduce severity, which is very important but we don't track whether flow limitations identified by the machine occurred for 1 minute or 8 hours. obviously that difference is significant. my flow limitations in terms of severity might be high but maybe they only occurred for 5 minutes. that might be less concerning than a lower level of severity that occurs all night long. we can kind of see that in the graph but we can see apnea events too. comparison is made easier by indexing.

so the essential purpose would be to enable comparisons between nights and more importantly between settings. a quick way to determine if a setting change helped or hindered treatment of flow limitations. for example, similar to above, maybe my setting change had no apparent effect on severity but afterwards the 'density' of my flow limitations rose from 25% of usage hours 2 nights ago to 39% last night. that strikes me as useful information. or maybe under my clinic's settings, my machine is reporting more that 50% of my usage hours are flow limited. we don't know (yet) what thresholds are significant, but that's still good info to take back to my sleep doc.

I don't know much about this kind of stuff so I'm winging the details and am counting on you much smarter members to consider how to implement it - if there's interest in the general idea.
Post Reply Post Reply
#14
RE: flow limitations expressed as an index?
I should be more detailed. I have a hard time realizing that everyone can't read my thoughts.

My question is; What would we do differently when using this data for treatment? Also, it is very difficult to put the analysis we make with our mind and eyes into abstract values. Case in point; How many years has it taken for computers to attempt image recognition?
Crimson Nape
Apnea Board Moderator
www.ApneaBoard.com
___________________________________
Useful Links -or- When All Else Fails:
The Guide to Understanding OSCAR
OSCAR Chart Organization
Attaching Images and Files on Apnea Board
Apnea Helpful Tips

INFORMATION ON APNEA BOARD FORUMS OR ON APNEABOARD.COM SHOULD NOT BE CONSIDERED AS MEDICAL ADVICE. ALWAYS SEEK THE ADVICE OF A PHYSICIAN BEFORE SEEKING TREATMENT FOR MEDICAL CONDITIONS, INCLUDING SLEEP APNEA. INFORMATION POSTED ON THE APNEA BOARD WEB SITE AND FORUMS ARE PERSONAL OPINION ONLY AND NOT NECESSARILY A STATEMENT OF FACT.
Post Reply Post Reply






Donate to Apnea Board  
#15
RE: flow limitations expressed as an index?
"What would we do differently when using this data for treatment?"

again, I think I've already addressed this as best I'm able. see the examples I gave above.

current assessment of severity lacks important context; i.e., some measure of quantity over time. so the main difference is adding frequency, duration or both to severity. separately for the sake of simplicity.

initially I wouldn't expect treatment to change. what it would do is provide additional detail/context (what I'm calling density), in addition to severity, to help understand whether treatment of fl is desirable/necessary and for easy comparison to help understand whether treatment actions are effective.

high level severity for just a few minutes may not be terribly significant. low level severity over a longer time may be. to my knowledge, other than a subjective visual assessment of the graph, there is no reporting of this temporal component.

seems to me there are at least the following components of flow limitations: severity however it's currently determined, severity if/as can be determined someday by the shape of the waveform (with a nod to 2Sleepbetta and others, ideally, understanding that would someday result in new improved type- or shape- specific treatments), and density which is frequency, duration or both.

while it could get complicated, what I'm suggesting as a starting point is actually quite simple. I'm not a coder/developer but it looks like density is relatively easy to figure and report using available data.

thanks for considering the suggestion. I've reached the limit of my motivation and ability to promote/defend this it so I won't waste any more of anyone's time on it.
Post Reply Post Reply
#16
RE: flow limitations expressed as an index?
Please don't stop on my account. I was just trying to get my mind around your idea. When it isn't your idea, it can be hard to grasp the other fellow's concept path.
Crimson Nape
Apnea Board Moderator
www.ApneaBoard.com
___________________________________
Useful Links -or- When All Else Fails:
The Guide to Understanding OSCAR
OSCAR Chart Organization
Attaching Images and Files on Apnea Board
Apnea Helpful Tips

INFORMATION ON APNEA BOARD FORUMS OR ON APNEABOARD.COM SHOULD NOT BE CONSIDERED AS MEDICAL ADVICE. ALWAYS SEEK THE ADVICE OF A PHYSICIAN BEFORE SEEKING TREATMENT FOR MEDICAL CONDITIONS, INCLUDING SLEEP APNEA. INFORMATION POSTED ON THE APNEA BOARD WEB SITE AND FORUMS ARE PERSONAL OPINION ONLY AND NOT NECESSARILY A STATEMENT OF FACT.
Post Reply Post Reply
#17
RE: flow limitations expressed as an index?
I understand. I just don't think I have more to add. either there's interest or there isn't. the membership might have related ideas. and it's up to the developers to determine if it's do-able.
Post Reply Post Reply






Donate to Apnea Board  
#18
RE: flow limitations expressed as an index?
Sheepless, I didn't see this thread before -- but it sounds like you've been thinking about some of the things I've been messing with. My FL knowledge is ResMed-centric, too, as that's all I have.

My complaints start with the ResMed FL reporting being kind of haphazard. What I see wrong fits into three broad categories:

1) The machine logs a zero flow limitation with a timestamp, and right there at the timestamp those breaths look pretty nicely rounded. But then there is a several-minute gap where the machine doesn't log any new value for FL, so OSCAR reports that the zero is continuing, but all sorts of stuff is happening, including events and near-events, wild recovery breathing after events, snores, etc. In my data, most of the OAs are coinciding with one of these zero flow limit regions. I'm left looking at OSCAR and muttering, "dude! a 45-second obstructive apnea is 45 seconds of 100% flow limitation, not an absence of FL!". One sure tell that this zero is not a real zero is that the machine is raising the pressure in reaction to what it's seeing. So the zero in the time bracket is because the machine didn't need to do any sophisticated analysis of the wave form to react, because it already knew you were in trouble. So it didn't do the flow limit math so it didn't log the math it didn't do.

2) The machine logs zero as the FL value even though you can see the distortions of the inspiratory curve because the algorithm fails to recognize the FL that you can see. The tell there is that the machine drops the pressure because it has missed the FL.

3) The flow rate waveform is gyrating around wildly (indicating a snore, recovery breathing, an arousal, or some combination) and the concept of the subtle deformations of the waveform that indicate flow limits make no sense at all. So the machine is just punting when it leaves the value the same as it was when it could calculate a value.

#2 is hard, but I think that the first and third cases ought to be fairly straightforward to do some level of processing of the data to get some new more sophisticated metric. Like first of all every Obstructive Apnea has a beginning and ending time (well, an end time and a duration, but that's the same thing). So my first pass would be to set the FL to 1.00 for the entirety of each OA. Second pass would be to look for regions of 0.0 FL where the machine is raising the pressure. Maybe looking at the changes in tidal volumes, insp time vs exp time, etc. might give us a region to score some level of FL where we know there's a false "hangover zero". Finally, if the waveform is wildly erratic and we can't figure out a more plausible value but know zero is silly, I would set the FL to "no data" for the duration of that.

All in all, I think that if you created an alternate post-processed FL data stream and then ran statistics on that then those median and 95% levels would really give you a better picture of what's going on. Those areas counted as zero now that I would argue are NDs -- those are heavily skewing the median to far lower than it really is.
Post Reply Post Reply
#19
RE: flow limitations expressed as an index?
(03-30-2021, 01:18 PM)sheepless Wrote: I wonder if it would be possible, and if others' think it worth considering, to express flow limitations as an index in addition to the measure of 'severity' we see now.
In one word: you can't.  The term "flow limitation" is somewhat misleading: it may refer to an actual event which occurs during breathing.

But this is not what's shown in OSCARS diagram labeled "Flow Limitation" as a number between 0 and 1 and should better be called "Flow Grading" as it's nothing more or less then an evaluation of your breathing which is continuously monitored.  Here "0" corresponds to an "A" and "1" to an "F".  As the Flow Grading is an ordinal value, no indexing, not even means, are allowed to calculate, only quantiles -- but what for?! Furthermore, that "Flow Limitation" often doesn't refer to a physical event as, for example, an obstruction may happen with an indicated "flow limitation" of zero ...

A very complex algorithm -- each manufacturer has its own algorithm by the way -- evaluates the flows form, change  and much more resulting is a more or less arbitrary number between 0 and 1, corresponding grade of your breathing's "quality".  Only a well trained eye may the judge  which physical event, the "real" flow limitation, corresponded to the given flow shape.  Here's a link to ResMed's algorithm: https://patentimages.storage.googleapis....5662A1.pdf  They even calculate (p. 15) some mumbo jumbo as  "the Shannon entropy of the normalized snore vector" as well as a "Persisting Flattening index" to get the "fuzzy persistent flattening" and much more wizardry.

From here the "Flow Grade" makes only sense on an APAP machine, where the pressure is adjusted to that grade accordingly, being somewhat "blind" to the "real" physical events.  But who cares: it works mostly (to our pleasure)!

Hence don't overestimate the so called "Flow Limitations" shown by OSCAR.  A decent look at the Flow Rate discloses much more about your breathing.
Post Reply Post Reply
#20
RE: flow limitations expressed as an index?
I would add an even more basic caveat... At least with the ResMed machines, the flow limit reporting is a series of timestamps and value pairs. You can see this in OSACAR because the graph is a series of stair steps with horizontal lines across the gaps between the vertical lines which are the only places where the data is reported. 

What is obvious from the data is that the ResMed algorithm is somewhat haphazard about when it chooses to write a timeStamp/flowLimit pair out to the card. So all of those horizontal lines? Except for the left-hand end of each one, that’s not even ResMed-reported data. 

I’m thinking about doing a deep dive into the code and seeing if I can produce an alternate graph which plots the data with just the reported dots and not the lines between them... (Or maybe I’ll use DataGraph since that’s a tool I’m a little more familiar with than qt)
Post Reply Post Reply






Donate to Apnea Board  


Possibly Related Threads...
Thread Author Replies Views Last Post
  Blank flow rate in Oscar pappingit 7 58 05-02-2021, 12:58 PM
Last Post: sawinglogz
  Flow Limitation Units in Statistics cathyf 15 410 04-01-2021, 02:10 AM
Last Post: cathyf
  Oscar: Weird spikes in flow rate and other feedback readings MikeBob 5 228 03-05-2021, 07:52 PM
Last Post: MikeBob
  Sleepyhead using Icon auto / Flow chart John McGowan 10 818 02-22-2021, 02:46 AM
Last Post: GuyScharf
  PS BIPAP ASV not showing Flow Rate in Oscar 19MaleMabyeUARS? 1 164 01-06-2021, 10:59 PM
Last Post: Sleeprider
  Flow rate graph shirsh98 9 329 12-29-2020, 02:32 PM
Last Post: Sleeprider
  reading the flow chart staceyburke 7 443 11-19-2020, 09:21 AM
Last Post: multicast


New Posts   Today's Posts






About Apnea Board

Apnea Board is an educational web site designed to empower Sleep Apnea patients.