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

Call for Excel VBA help: to support effort to clarify/understand FL
#11
RE: Call for Excel VBA help: to support effort to clarify/understand FL
fyi, I for one appreciate your efforts and updates. I hope you get some useful feedback. unfortunately, I have nothing to offer but moral support as this level of detail is way beyond me.
  Shy   I have no particular qualifications or expertise with respect to the apnea/cpap/sleep related content of my posts beyond my own user experiences and what I've learned from others on this site. Each of us bears the burden of evaluating the validity and applicability of what we read here before acting on it.  
 
Post Reply Post Reply






Donate to Apnea Board  
#12
RE: Call for Excel VBA help: to support effort to clarify/understand FL
An update that shows a bit more progress zig-zagging toward a better understanding of what ResMed flow limit (FL) flags can or do reflect in my own breathing. There are few surprises in the graphs, but the varying volumes do show applicabilty of FL flags while suggesting other factors Resmed patents and algorithm cover.

In a nutshell the alternative approach being taken is to present examples of inspiratory volume reductions in relationship both to flow limitations (lower case "fl", those limitations below the Resmed FL flag threshold level) and to those flagged limitations above that minimum threshold. My present comments about the graphs are only three. 1) M-tipped inspirations are accompanied, as is usual for me, by a trailing FL flag that starts late in that breath's exhalation; moreover, there are two successeive M-tips which present a stacked FL with a seond step. Yes, there is a flow drop there but I believe the M-tip gets special Resmed attention to shape and flagging; 2) A few tiny cardiogenic flows show along the 0-axes; 3) I'm curious why, at about 09:47:25, there is no FL as there is at 09:47:00 (I see the former form frequently and without FL. The algorithm may "see" it as breath-holding?

I show in the attachment's top-most graph a crude representation of the loss of Tidal Volume (TV) due both to restriction from flow limitations and from "false-losses" that arise from natural/normal/usual breathing variations--simple changes of respiratory rate and other sleep metrics that vary during sleep. Accordingly, the baseline in the upper graph arbitrarily presents drop bars from a fictitious constant baseline, one chosen to be near a fictitious steady Miinute Volume (MV) of 5 to 7 liters/min. In other words, if the "drops baseline" is set at 7 L/min for TV and the breathing is actually 4 L/min then the drop would be 7 minus 4 L/min--it would be 3 L/min. Unlike the ease of showing real TV from the zero baseline in the data, there is no fixed volume baseline for computing inspiratory loss.  

In theory--and as could have been presented much more clearly--the upper graph and the third graph down should have the same vertical scale that would mate together along their bottom and top edges respectively like two pieces of paper would match after tearing the sheet in two.

Three side notes: 1) Re the bottom attachment in post no. 10 above: some days after posting, it struck me that the seeming absence of flow limitations (according to the cruder approach there) was likely due to oblivious  filtering out of  the largest flow limitations which would be expected there, given what SleepyHead presented as FL. I intend to check that out as I blunder along this path. 2) Those numbers in the 16 thousands along my graphs' axes in the attachment are the seconds past midnight. It is a big hassle to deal with conversion of them--as they are used in files underlying OSCAR presentations to usual clock numbers. They could be divided by 86400 (seconds in a day) and then be formatted as time values in Excel if I weren't too lazy or if space for presenting clock time wouldn't be crowded when showing and dealing with logged times in fractions of seconds. The graphs are closely, if not perfectly, synchronized, though crudely presented.

Readers' constructive and provocative observations would be helpful.

   
I have no particular qualifications or expertise with respect to the apnea/cpap/sleep related content of my posts beyond my own user experiences and what I've learned from others on this site. Each of us bears the burden of evaluating the validity and applicability of what we read here before acting on it.  (Disclaimer use permitted by sheepless)

 
Post Reply Post Reply
#13
RE: Call for Excel VBA help: to support effort to clarify/understand FL
I have to admit this is so far over my head right now, but good luck with your continued research and I hope that (one day) I'll have some informed opinions to share. Smile


See my comparison of Viatom/Wellue and CMS50F oximeters here.

Not a doctor, definitely not your doctor, all advice is given as-is and represents simply my own understanding as a fellow patient and OSCAR user.
Post Reply Post Reply
#14
RE: Call for Excel VBA help: to support effort to clarify/understand FL
2SleepBetta, this is over my head, too, at this point, but my husband the theoretical physicist says that these time-series data analyses that we are doing are inherently the same sorts of things he does in his research.

If you have access to a Mac, there is a tool called DataGraph which he says is the hands-down best tool for doing this kind of analysis.
Post Reply Post Reply






Donate to Apnea Board  
#15
RE: Call for Excel VBA help: to support effort to clarify/understand FL
(06-18-2021, 12:13 PM)Ratchick Wrote: I have to admit this is so far over my head right now, but good luck with your continued research and I hope that (one day) I'll have some informed opinions to share. Smile

Oops, Ratchick, this thank you and the followup post below are far overdue. I believe I thanked others for coming by and helping. Please excuse me. If it is over your head it is only because I have not presented the matter well, a fact in any case, and this post, dang it, will be no improvement. Hammer  My effort to understand the Resmed FL flag continues.

This post and its attachments continue my post #10 above which has fuller explanation I do not restate here. The second graphic below is from post #10. Here I focus on what still seems surprising to me, if not anomalous, in the FL during the whole more extended period here (0430-0530 or 0530-0630 depending on the time zone used before my copying the upper graph). One can see the then-SleepyHead montage show large FL, but the individual inspired volumes do not seem to be reduced concordantly. To me a drop in inspired volume is a true witness to some kind of airflow limitation and, no, that drop is not expected to be totally consistent with the Resmed FL indications which reflect the algorithm's consideration of not only volume reduction, but also shapes in general, the flatness shape and the respiratory rate. I think the graphic shows a  larger impact those other factors have in displays of FL flags.

I'd appreciate anyone's analytical comments that would help me and others understand what I see in the  attached curves. My new thought, getting back to this--barring blunders in the graphics--is that the seemingly larger than proportionate FL flagging reflects shape, RR and flatness and their changes to a greater extent than I expected. Inspired volume  still is a major factor, but it seems to be secondary and have lowered impact at times. For simplicity, the shortfall graphs assume a (fictitiously) normal inspiration is 0.5 L.

I expect to think and comment further, about the graphs, but need to shut down. First, a few words about the three uppermost graphs. The top and bottom ones are duplicates, the top one is only to show that it would drop down and fit like a puzzle piece into the middle graph. The middle graph depicts inspiratory volume alone for each single breath, as determined by numerical integration of the flow rate curve--the adding together of .04" wide strips of inspired volume at each rise and drop along the flow rate curve for each inspiration (all that above the zero-axis). The bottom duplicate is there--for no good reason, mainly to be next to the Resmed FL indicators.

   

   
I have no particular qualifications or expertise with respect to the apnea/cpap/sleep related content of my posts beyond my own user experiences and what I've learned from others on this site. Each of us bears the burden of evaluating the validity and applicability of what we read here before acting on it.  (Disclaimer use permitted by sheepless)

 
Post Reply Post Reply
#16
RE: Call for Excel VBA help: to support effort to clarify/understand FL
2SB -- I started with this screenshot

http://www.apneaboard.com/forums/attachm...?aid=34055
It's from a period around 5:20 on jun 19 where there was an extended amount of time with a flow limit at about .5, that was followed by an arousal and then non-flow limited breaths for awhile. I made screenshots on either side of that one and I put them all together in a subfolder.

This is a folder of a bunch of screenshots from jun 19, and also a csv of the data dump for the day.
https://www.dropbox.com/sh/llfz3rbdtkcd8...1oAPa?dl=0

Inside that folder is a subfolder named 5.20 which contains those six screenshots around 5:20. (those are full OSCAR window screenshots, with all of the channels.)

I don't know if you can make anything of that...

Is there any way to get OSCAR to dump out the data in the flow rate curve? This
https://www.dropbox.com/sh/qku3ksg12xi9s...dFYRa?dl=0
is a link to my DATALOG/20210619 file from my data card -- not sure if it's possible to get the values for the flow rate curve over that region -- it might help to do more accurate calculations if you had the numbers and weren't trying to measure them off the pictures.

So when I look at that graph and compare the flow-limited with the not-flow-limited sides, it looks like the respiration rate is the same, and the minute vent is not really lower during the flow limit. Just eyeballing it, the flow-limited breaths are flatter and don't go up as high and are more squared off, while the not-flow-limited are taller and pointier and more like a triangle. Without measuring it carefully, I think that it looks like the area under those curves is the same on both sides -- which I think is what is meant by increasing the duty cycle?
Post Reply Post Reply
#17
RE: Call for Excel VBA help: to support effort to clarify/understand FL
Long post as always: The main point is in the graphic with its summary, attached below, which shows that the flow limitation flags (FL) in the sleep sample are slightly less more than half explained by drops in inspired volumes. As discussed regularly, three other metrics and indices are factored into machine "decisions" to show and describe the FL flag with duration and severity indications.

The attachment continues from  analysis of the set of FR and FL curves and TV drops as discussed beginning  at Post #10 above. There a question was raised about the time period that was marked with a green frame around it. It seemed there was a mismatch of large FL versus the absence of the expected drops in inspired volumes.
  
Subsequent posts and the graph below continue that theme. Below I see strong but not conclusive evidence that the first --surprising-to-me--impression was  correct. FL depicted below are not just about drops in inspiratory volume, but almost half of the graph features noted were pairs of FL and TV drops.
  
The summary chart with percentages was prepared by rating prominences, and lacks thereof,  of both FL and TV drops. Absences of FL and of TV and presences of large FL with sharp drops were roughly scored as 6's using a scale of 1 to 6, 6 being the highest level of agreement for FL and expected drop and 1 being the lowest level of agreement.

Post #8 presented a different short period from a with FR, FL and a TV-drop curve. The focus was on a scattered few distinct FL markers with short durations of, say, 8-12 seconds and low severity. My guess is that there  the lower density and smaller  variations among  the 4 drivers of FL flagging might "strain or tax" the algorithm, which governs pressure changes, less than in the one hour case below. Such dense patterns may be more problematic for the algorithms, IMO.

My take away to date: Indications are that the inscrutable wiggles, the "goat entrails" and the tea leaves we can sometimes read to our therapy's advantage--all those are hints in the inspiratory part of our FR curves and of our mask pressure curves (when expanded). The changes in FR wave peaks are due to our breathing irregularity, some of that is benign, some not. Sensing them is quite important for effective performance of the Resmed devices. But like baseball umpires, a few bad calls are a part of the game.

That said, it isn't at all simple. Many with low AHI struggle and labor against nearly continuous air flow restrictions which cause arousals and unrestful sleep. Their air flows may or may not present their device with enough variation in restrictive effects--enough changes--to trigger their device's adjustments of pressures. There may only be a few indications (or none) of flow limitations and no FL flags. Their wave shapes may look normal or close to it.

Leaders here are teaching us , among other things, to look at the length of inspiration times in those cases, those times versus exhalation times. The patient is working harder and longer to draw a needed breath--an important clue. 

So much to learn! Like peeling an onion. It's layer after layer and knowing where or whether to continue or to quit an approach and try different therapy settings and devices.





   
I have no particular qualifications or expertise with respect to the apnea/cpap/sleep related content of my posts beyond my own user experiences and what I've learned from others on this site. Each of us bears the burden of evaluating the validity and applicability of what we read here before acting on it.  (Disclaimer use permitted by sheepless)

 
Post Reply Post Reply






Donate to Apnea Board  
#18
RE: Call for Excel VBA help: to support effort to clarify/understand FL
I have a few dumb questions, not likely to further your exploration but might help me understand what you're doing, at least a little.

what are the red downward lines or waveform below the 0 line in the volume graph, also mirrored at top of graph at 1.5?

oscar reports tidal volumes from the med column and up from roughly 300 to 1000 ml. what does your scale of 0 to 2.5 represent?

what is the gray area behind(?) between(?) the black flow rate waveform? not (necessarily) the area that appears to be a leak but the gray that runs the length of the flow rate (where I'd expect to see white space)?

is this non-compact, choppy or hairbrushy flow rate waveform common for you? is this pattern caused by flow limitations?

digging into this stuff and gaining a better understanding of the meaning of flow limitation and how it's determined is a noble exercise and interesting in itself, but do you also have a sense of how this knowledge might someday influence, say, optimizing titration? or more generally, where do you think / hope this knowledge might take us?

not trying to pin you down and 'I don't know' is an acceptable response. just hoping to understand your interesting posts on the subject.
  Shy   I have no particular qualifications or expertise with respect to the apnea/cpap/sleep related content of my posts beyond my own user experiences and what I've learned from others on this site. Each of us bears the burden of evaluating the validity and applicability of what we read here before acting on it.  
 
Post Reply Post Reply
#19
RE: Call for Excel VBA help: to support effort to clarify/understand FL
Many thanks and its good to see you posting after an absence, Sheepless. My responses to your  comments, in black font below, are in my blue-colored font, I hope, but can scarcely make that out on my monitor, set as it is.

I have a few dumb questions, not likely to further your exploration but might help me understand what you're doing, at least a little. Necessary questions eliciting, among other things, corrections. Not dumb.

what are the red downward lines or waveform below the 0 line in the volume graph, also mirrored at top of graph at 1.5? The red line, if its length were added to the length of the corresponding "hairbrush bristle" above it should make a bar, there (and everywhere else), that is  0.5 L long/tall overall, that 0.5L level represents the assumed inspiratory flow volume of just one average adult-human inspiratory wave at that same time. You see in the bristle heights my breathing variation, that night, from my (helplessly assumed) 0.5L/breath average that I based the uppermost graph and red bar graphs on. 

In a manner of speaking, the red bars and uppermost descending bars identically represent the upper part of all individual inspiration volumes that are "missing" for whatever flow limiting reasons. It is just a method, the only one I could think of, to show what a FL flag means relative to, er, an actual local flow limitation or actual flow-volume reduction in other words.

Yeah, the 2.5 is most confusing indeed, along with my mislabeling the units on the vertical axis as "L/min" when it should be "L/breath". The 2.5 choice was an artifice that would help me position the entire set of uppermost waves so they, collectively, could be seen more clearly as one nearly perfect matching puzzle piece, relative to the entire hairbrush graphic below. The red bars, I saw, would help with comparisons of missing volumes to FL flags just below.

oscar reports tidal volumes from the med column and up from roughly 300 to 1000 ml. what does your scale of 0 to 2.5 represent? I think my response above covers that.

what is the gray area behind(?) between(?) the black flow rate waveform? not (necessarily) the area that appears to be a leak but the gray that runs the length of the flow rate (where I'd expect to see white space)?

is this non-compact, choppy or hairbrushy flow rate waveform common for you? is this pattern caused by flow limitations? The brush like graph represents each individual inspiratory wave volume for the 1-hour period. Each separate vertical bar height ("bar" I call them) varies up and down, mostly near 0.5L/breath, the "adult human average" which I assumed-lacking my appropriate, personal single wave or sliding window average volume for my local inspiratory wave volumes.

digging into this stuff and gaining a better understanding of the meaning of flow limitation and how it's determined is a noble exercise and interesting in itself, but do you also have a sense of how this knowledge might someday influence, say, optimizing titration? or more generally, where do you think / hope this knowledge might take us?  I think that for my nerdy type the takeaway is to remember, when seeing high levels of high FL that, based on my sample, I should  look for help interpreting FL flags using either/or-both of the following, (1), by examining greatly expanded views of  deformations of FR peaks and by examining Inspiration and Expiration times, (2)  by doing the work of generating a graph showing my losses of inspiratory volumes, drops in TV. As I type this, I realize need to revisit the nature of the TV curve we see in Oscar, how frequently are its data sampled, for one most important thing. Maybe my whole exercise here is pointless. But as I remember, it has been difficult to see in the TV curve the short duration drops in TV, though I still keep it in view.

not trying to pin you down and 'I don't know' is an acceptable response. just hoping to understand your interesting posts on the subject. Thanks again for reminding me, with my acknowledged huge deficiencies in communication skills, of McCluhan's? ...?? No (click following link) Illusion of having communicated- -looking at the presenter's side of communication duties.

A final edit I hope: It is confusing why I write of breath volumes in these posts. A flow rates are not volume. The rate can be high, low, short or long. The problem is that flow rates must be time weighted in order to score/add-up/determine the relative ventilation sufficiency (and TV's ) of individual breaths, of pairs and of larger numbers of inspirations and then compare those results to TV drops at FL along the FR (rate) curve. Done properly, as I believe I have but poorly explained, I can better know the limitations and benefits of the FL flag, after, lo, the past 5 years of not catching on. 

I'd be remiss in failing to mention that I believe member multicast recently wrote that FL were half determined by inspiratory wave shape. My work and sample seem to confirm that.
I have no particular qualifications or expertise with respect to the apnea/cpap/sleep related content of my posts beyond my own user experiences and what I've learned from others on this site. Each of us bears the burden of evaluating the validity and applicability of what we read here before acting on it.  (Disclaimer use permitted by sheepless)

 
Post Reply Post Reply
#20
RE: Call for Excel VBA help: to support effort to clarify/understand FL
The link about communication was lost in and edit: it was https://quoteinvestigator.com/2014/08/31/illusion/ and I could not hot link it in an additonal quick edit.


The hotlink: https://quoteinvestigator.com/2014/08/31/illusion/
I have no particular qualifications or expertise with respect to the apnea/cpap/sleep related content of my posts beyond my own user experiences and what I've learned from others on this site. Each of us bears the burden of evaluating the validity and applicability of what we read here before acting on it.  (Disclaimer use permitted by sheepless)

 
Post Reply Post Reply






Donate to Apnea Board  


Possibly Related Threads...
Thread Author Replies Views Last Post
  OSCAR Support for DSPD (Time/Session Issue greenwsj 7 232 09-02-2021, 02:45 PM
Last Post: fakename1845
Smile Oscar support for ResMed AirSense 11 Autoset tmiller 4 494 08-23-2021, 05:08 AM
Last Post: tmiller
  RESSCAN - PURE VIRTUAL FUNCTION CALL [see post #2 for interim solution] srlevine1 4 176 07-29-2021, 09:59 AM
Last Post: srlevine1
  ResMed iVAPS support? alshayed 4 142 07-12-2021, 04:46 PM
Last Post: alshayed
  Support for Fisher & Paykel ICON+ Auto on SleepyHead or Oscar JaX F 2 269 04-05-2021, 12:17 PM
Last Post: A KLERK
  Help with Excel Correlation please SevereApnea 5 298 03-29-2021, 09:26 AM
Last Post: Crimson Nape
  Adding Daylight Savings support to OSCAR greyham 16 752 03-25-2021, 05:23 PM
Last Post: kappa


New Posts   Today's Posts






About Apnea Board

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