Interpreting Flow Graphs

General Discussion on any topic relating to CPAP and/or Sleep Apnea.
Rick007
Posts: 50
Joined: Sun Oct 20, 2013 4:40 pm
Location: Ontario, Canada

Interpreting Flow Graphs

Post by Rick007 » Thu Nov 21, 2013 12:05 pm

I have attached an image of the Flow Graphs from both SleepyHead and Rescan for the same period of time. Although SleepyHead offers an option to change the Y-Axis scale, it doesn't work for me, so I changed the scale on the Rescan graph so that both waveforms were as similar as possible.

You can see that the flow values on Rescan range from a high of +30 L/Min to a low of -40 L/Min, for a total range of 70 . The SleepyHead graph ranges from 0 to -82.5, for a total range of 82.5. The SleepyHead graph does not specify units, so perhaps they are not the same as Rescan.

How can I explain the difference between the two programs. I am using a F&P Eson nasal mask, and I have programed my Resmed S9 ELite with Mask Type=Nasal.

Could it be that SleepyHead is accounting for the mask's internal leakage and Rescan isnt. I found a setting in SleepyHead where you enter the mask type, but making changes here does not change the Flow graph at all.

Are the values of the flow graph important in determining therapy, or is just the shape of the waveform important for determining Flow Limitations, etc.

I had a theory that the minimum to maximum swings on the flow graph would indicate the overall volume of air inhaled. Based on that theory I compared the Minute Vent graphs from both programs and I found the values are much closer between programs.

Image

Image

User avatar
Pugsy
Posts: 65253
Joined: Thu May 14, 2009 9:31 am
Location: Missouri, USA

Re: Interpreting Flow Graphs

Post by Pugsy » Thu Nov 21, 2013 12:15 pm

I don't know if removing the check mark from "use pixmap caching" will fix the SleepyHead graph or not but worth a try...it fixes a lot of other stuff.
Preferences...Appearance tab...remove the check mark and restart SleepyHead to see if it helps.

Otherwise I have no idea what is going on and will defer to hopefully someone that does.
You may have stumbled across a SleepyHead bug. The way that ResScan presents the flow graph is the normally seen way.

I just now looked back on my old S9 reports...the flow rate graph is all below zero but my PR S1 flow rates are as they should be...some above 0 and some below 0.
And since I have "use pixmap caching" unchecked...I doubt it will fix your issue but won't hurt to try.

My vote is for a SleepyHead graph bug. I no longer have S9 data to try the new unstable build to see if that makes any difference in the new build. I know Mark was working on some axis stuff but don't know exactly what.

_________________
Machine: AirCurve™ 10 VAuto BiLevel Machine with HumidAir™ Heated Humidifier
Additional Comments: Mask Bleep Eclipse https://bleepsleep.com/the-eclipse/
I may have to RISE but I refuse to SHINE.

Rick007
Posts: 50
Joined: Sun Oct 20, 2013 4:40 pm
Location: Ontario, Canada

Re: Interpreting Flow Graphs

Post by Rick007 » Thu Nov 21, 2013 1:42 pm

Pugsy wrote:I don't know if removing the check mark from "use pixmap caching" will fix the SleepyHead graph or not but worth a try...it fixes a lot of other stuff.
Preferences...Appearance tab...remove the check mark and restart SleepyHead to see if it helps.
Thanks for the suggestion, but the graphs look the same with Pixmap Caching either on or off. I also tried to run the nightly unstable version, and discovered a bug during data input where there is no option to "choose" data. I will attempt to notify the developer via the Sourceforge site.

User avatar
jaybeem
Posts: 189
Joined: Sat May 04, 2013 7:55 pm
Location: CA, USA

Re: Interpreting Flow Graphs

Post by jaybeem » Thu Nov 21, 2013 2:14 pm

This is exactly the way my flow looks...and has since May. I don't think this pattern is a bug. Could it be restricted flow? I have been considering posting mine. My numbers look great. I only awaken once a night, always during my first REM stage. But my sleep continues to be long and not as restful as I would like.

How is the quality of your sleep?

_________________
Mask: Nuance & Nuance Pro Nasal Pillow CPAP Mask with Gel Nasal Pillows
Humidifier: S9™ Series H5i™ Heated Humidifier with Climate Control
Additional Comments: Pressure 8. EPR 1. Zero ramp. SleepyHead software.

Rick007
Posts: 50
Joined: Sun Oct 20, 2013 4:40 pm
Location: Ontario, Canada

Re: Interpreting Flow Graphs

Post by Rick007 » Thu Nov 21, 2013 2:39 pm

jaybeem wrote:This is exactly the way my flow looks...and has since May. I don't think this pattern is a bug. Could it be restricted flow? I have been considering posting mine. My numbers look great. I only awaken once a night, always during my first REM stage. But my sleep continues to be long and not as restful as I would like.

How is the quality of your sleep?
The waveforms in my original post do look fairly normal (as far as I can tell). There are other times during the night that the peaks become flattened, and this coincides with high levels on the Flow Limitations graph. Overall I don't notice any change in my sleep quality during the few weeks that I have been on PAP therapy.

My original question wasn't about the shape of the waveforms, but rather why the ResScan software showed different values from SleepyHead. In the SleepyHead graphs, all the flow is negative in value. By definition that would mean that my S9 Elite was not pushing any air into my lungs, and I know that just isn't so.

User avatar
jaybeem
Posts: 189
Joined: Sat May 04, 2013 7:55 pm
Location: CA, USA

Re: Interpreting Flow Graphs

Post by jaybeem » Thu Nov 21, 2013 5:24 pm

I though "good" flow limitation graphs were supposed to have an almost bell-curve shape with rounded tops. What does "good" or "normal" look like?

_________________
Mask: Nuance & Nuance Pro Nasal Pillow CPAP Mask with Gel Nasal Pillows
Humidifier: S9™ Series H5i™ Heated Humidifier with Climate Control
Additional Comments: Pressure 8. EPR 1. Zero ramp. SleepyHead software.

User avatar
Pugsy
Posts: 65253
Joined: Thu May 14, 2009 9:31 am
Location: Missouri, USA

Re: Interpreting Flow Graphs

Post by Pugsy » Thu Nov 21, 2013 6:02 pm

Rick007 wrote:My original question wasn't about the shape of the waveforms, but rather why the ResScan software showed different values from SleepyHead. In the SleepyHead graphs, all the flow is negative in value. By definition that would mean that my S9 Elite was not pushing any air into my lungs, and I know that just isn't so.
It's a SleepyHead S9 bug. The air really isn't under enough pressure to be "forced" into your lungs. The main function of the cpap pressure is simply to hold the airway open and prevent the collapse of the airway tissues. At 20 cm pressure the machine can't even inflate a balloon.
jaybeem wrote:I though "good" flow limitation graphs were supposed to have an almost bell-curve shape with rounded tops. What does "good" or "normal" look like?
Not sure if this has enough examples for you to look at but check it out.
http://www.resmed.com/us/assets/documen ... er_eng.pdf

and this might help

Image

_________________
Machine: AirCurve™ 10 VAuto BiLevel Machine with HumidAir™ Heated Humidifier
Additional Comments: Mask Bleep Eclipse https://bleepsleep.com/the-eclipse/
I may have to RISE but I refuse to SHINE.

User avatar
jaybeem
Posts: 189
Joined: Sat May 04, 2013 7:55 pm
Location: CA, USA

Re: Interpreting Flow Graphs

Post by jaybeem » Thu Nov 21, 2013 6:37 pm

Thanks, Pugsy! I don't see anything quite like Rick007's or my spike up/plateau/spike up to pointy top/spike down.

But I also have several episodes of "expiratory mouth breathing with normal inspiratory flow" episodes and now, thanks to you, know what those are.

_________________
Mask: Nuance & Nuance Pro Nasal Pillow CPAP Mask with Gel Nasal Pillows
Humidifier: S9™ Series H5i™ Heated Humidifier with Climate Control
Additional Comments: Pressure 8. EPR 1. Zero ramp. SleepyHead software.

User avatar
Todzo
Posts: 2014
Joined: Tue Apr 24, 2012 8:51 pm
Location: Washington State U.S.A.

Re: Interpreting Flow Graphs

Post by Todzo » Thu Nov 21, 2013 9:02 pm

Rick007 wrote:I have attached an image of the Flow Graphs from both SleepyHead and Rescan for the same period of time. Although SleepyHead offers an option to change the Y-Axis scale, it doesn't work for me, so I changed the scale on the Rescan graph so that both waveforms were as similar as possible.

You can see that the flow values on Rescan range from a high of +30 L/Min to a low of -40 L/Min, for a total range of 70 . The SleepyHead graph ranges from 0 to -82.5, for a total range of 82.5. The SleepyHead graph does not specify units, so perhaps they are not the same as Rescan.

How can I explain the difference between the two programs. I am using a F&P Eson nasal mask, and I have programed my Resmed S9 ELite with Mask Type=Nasal.
My best guess is that the differences result from two different rounding techniques or factors. The SleepyHead waveform looks less sharp in it's transitions so I suppose if we had the source code for both and could see the actual formula used to calculate the graphed transitions between the samples the reasons for the difference would become obvious.
Rick007 wrote:Could it be that SleepyHead is accounting for the mask's internal leakage and Rescan isnt. I found a setting in SleepyHead where you enter the mask type, but making changes here does not change the Flow graph at all.
This should not be a factor in graph shape although I have noted that excessive leakage will reduce the apparent volumes where it makes not sense that the volume would change. Leakage may play a part in this.

In general it looks like Rescan tries to determine what the flow rate into and out of the nose would be so zero occurs between the inhale and exhale tips.

It looks like SleepyHead simply displays the absolute value.
Rick007 wrote:Are the values of the flow graph important in determining therapy, or is just the shape of the waveform important for determining Flow Limitations, etc.
Absolute value does not interest me as much as how the flow changes over the course of a breath. Working with recorded audio I have learned to recognize when obstruction is likely to be indicated by how the graph changes. It is important to me to see indications of obstruction.

In general the width of the graph tends to show air volume used. That is important to me to help determine breathing stability.
Rick007 wrote:I had a theory that the minimum to maximum swings on the flow graph would indicate the overall volume of air inhaled.
The volume of air used is the area under the waveform (area taken up by the waveform relative to zero). So in the rescan graph volume inhaled and then volume exhaled are shown as the area traced by inhale and exhale above or below zero. Most use small rectangles evenly spaced on the graph with the edges poking out of the graph (and into the graph) with the height determined by a point on the graph taking a bit of distance each side to form the top of the rectangle with the bottom resting against the zero line and the width by the even distance each side. Add up all the little areas of the rectangles and you have the volume.
Rick007 wrote: Based on that theory I compared the Minute Vent graphs from both programs and I found the values are much closer between programs.
Minute volume I use to help me determine nighttime breathing stability. For me I like to see a value under 6 lpm for the nighttime average. I like to see the 95% no more than 3 lpm above that. I know that the max can vary for many reasons but it is comforting to see it close to the 95%. The higher the nighttime average and the further the others are above that the more I am concerned that breathing instability trounced my night of sleep.
May any shills trolls sockpuppets or astroturfers at cpaptalk.com be like chaff before the wind!

Rick007
Posts: 50
Joined: Sun Oct 20, 2013 4:40 pm
Location: Ontario, Canada

Re: Interpreting Flow Graphs

Post by Rick007 » Fri Nov 22, 2013 10:05 am

Hi Todzo, Lots of good information.
Todzo wrote:My best guess is that the differences result from two different rounding techniques or factors. The SleepyHead waveform looks less sharp in it's transitions so I suppose if we had the source code for both and could see the actual formula used to calculate the graphed transitions between the samples the reasons for the difference would become obvious.
I actually consider the waveforms very close in appearance, and I associate the slight differences due to the different X & Y scaling. The first pulse on the ResScan graph actually corresponds to the second pulse on the SleepyHead graph.
Todzo wrote:In general it looks like Rescan tries to determine what the flow rate into and out of the nose would be so zero occurs between the inhale and exhale tips. It looks like SleepyHead simply displays the absolute value.
That may be the case. I did note that the overall peak to trough values were not the same, but that difference may also be caused by the different Y-axis scale, and my interpolation of the values. I did attempt to look at the raw data from my S9 with a Hex Editor. I was hoping to determine if actual flow rates were recorded, or if just an offset from idle was recorded. Unfortunately there wasn't much data available in plain text to determine that.
Todzo wrote:Minute volume I use to help me determine nighttime breathing stability. For me I like to see a value under 6 lpm for the nighttime average. I like to see the 95% no more than 3 lpm above that. I know that the max can vary for many reasons but it is comforting to see it close to the 95%. The higher the nighttime average and the further the others are above that the more I am concerned that breathing instability trounced my night of sleep.
It is interesting that you arrived at those particular values. I understand that very low values would be a problem, but why would higher values be a problem?. Why are you concerned about staying under 6 lpm on average?. My median is 7.5 and 95% is 9.75.

Rick007
Posts: 50
Joined: Sun Oct 20, 2013 4:40 pm
Location: Ontario, Canada

Re: Interpreting Flow Graphs

Post by Rick007 » Fri Nov 22, 2013 10:13 am

Pugsy wrote:
jaybeem wrote:I though "good" flow limitation graphs were supposed to have an almost bell-curve shape with rounded tops. What does "good" or "normal" look like?
Not sure if this has enough examples for you to look at but check it out.
http://www.resmed.com/us/assets/documen ... er_eng.pdf
I wasn't sure if I was allowed to post images from the Resmed document, but the images on page 73 of the PDF document (linked to by Pugsy) show some great details of common waveforms including flow limitations which I seem to be suffering from. However I still can't find a pattern exactly like my own, with the plateau during the inhalation cycle.

User avatar
Todzo
Posts: 2014
Joined: Tue Apr 24, 2012 8:51 pm
Location: Washington State U.S.A.

Re: Interpreting Flow Graphs

Post by Todzo » Fri Nov 22, 2013 1:31 pm

Rick007 wrote:Hi Todzo, Lots of good information.
Todzo wrote:Minute volume I use to help me determine nighttime breathing stability. For me I like to see a value under 6 lpm for the nighttime average. I like to see the 95% no more than 3 lpm above that. I know that the max can vary for many reasons but it is comforting to see it close to the 95%. The higher the nighttime average and the further the others are above that the more I am concerned that breathing instability trounced my night of sleep.
It is interesting that you arrived at those particular values. I understand that very low values would be a problem, but why would higher values be a problem?. Why are you concerned about staying under 6 lpm on average?. My median is 7.5 and 95% is 9.75.
Post traumatic event the anniversary of the event would land me in ER for a couple of years. My therapy changed dramatically. I found myself waking up breathing very hard, very dry mouth, aerophagia. I totally lost circadian rhythm showing a non-24 sleep wake pattern with missing days. It was very hard on me.

Then, while listening to an Apnea related Webcast I was given some very basic instruction on how I could clear my nose simply by breathing less. It worked. It worked without me holding my nose and bobbing up and down (I wanted my hands free to take notes). This was my simple introduction to eucapnic breathing techniques.

As they talked about the physiology involved with hypocapnic, eucapnic, and hypercapnic breathing it became obvious to me that the panic attacks which landed me in ER and would keep me up for days at a time leading to the ER visit consisted of a whole lot of hyperventilation. A lot of hypocapnic breathing if you will. As I listened to the basics of how eucapnic breathing techniques work I became concerned that you could “go too far” and make for hypercapnic breathing if you did not have feedback of some sort. Well if you are hypocapnic vasoconstriction will occur and your heart will work hard to compensate. If you are hypercapnic vasodilation will occur and your heart will work hard to now try to fill the open hoses. It seems that lowest heart rate holding exertion constant might work as a guide.

So my pulse oximeter went on my wrist and I started the process sitting at my computer. It is not easy to control breathing rate and volume while watching for changes in heart rate to find the lowest heart rate with exertion held constant. It took me hours to get it down. Fortunately, along the way I noticed that the SpO2 reading would be consistent when the lowest heart rates were found. Back then an SpO2 reading of 96% at my desk (97% on a treadmill) resulted in the lowest heart rates (and a clear nose, and warm feet, and better thinking, more endurance, an un-knotted stomach (after several minutes) and no pain in my knees as I exercise and many other good changes I have found along the way). Likely because I moved to a higher altitude or perhaps because I am in better shape it is now 95% at the desk and 96% with constant moderate exercise. I make a point of using the pulse oximeter as a guide when I am on a cycle which is partially done to keep arthritis out of my knees. Pain in my knees does develop if I over breath (or at least it used to) and I like the idea of the tissues in the knee having the circulation and oxygenation they need to heal as I exercise.

Then it was time to try to do some pulse oximeter guided eucapnic breathing while using CPAP. I found this so hard to do that I decided to give up after only about an hour and practice my pulse oximeter guided eucapnic breathing for several days (I wore my pulse oximeter constantly for the first three months) to get it down a bit more before I would try it with the machine again. I know from this very well that the pressure of CPAP makes it difficult to breath eucapnically. For me it is already easy to breath too much. The pressure of CPAP simply adds to that problem.

So I hope you are beginning to understand why it is important to me to not over breath while using the machine. You would understand a bit better if you searched out what hyperventilation is and does. It would take you a bit further if you looked into the changes which occur in the brain due to it's vascular carbon dioxide reactivity. Further yet could you go if you looked into hypocapnic central apneas, CPAP and ventilatory instability and how this relates to respiratory effort related arousals. It is very very important to not over breath. But the pressure of CPAP makes it much easier to over breath.

As for why six liters per minute. I am average size and six liters per minute is the expected rate at rest. As well I know from looking at my data for several years now that I will feel and do better if I see things resting near six liters per minute average minute ventilation.

As for why I get concerned if the numbers not average go high, well, what do you think causes respiratory effort related arousals? And if you have one of those, your stress hormones will rise – so – you will be much more likely to have another. More events (of any kind) less breathing stability.
May any shills trolls sockpuppets or astroturfers at cpaptalk.com be like chaff before the wind!

Rick007
Posts: 50
Joined: Sun Oct 20, 2013 4:40 pm
Location: Ontario, Canada

Re: Interpreting Flow Graphs

Post by Rick007 » Fri Nov 22, 2013 2:47 pm

Todzo wrote:Post traumatic event the anniversary of the event would land me in ER for a couple of years. My therapy changed dramatically. I found myself waking up breathing very hard, very dry mouth, aerophagia. I totally lost circadian rhythm showing a non-24 sleep wake pattern with missing days. It was very hard on me.
That is a lot of information to digest quickly, but I will do some research into what you described. Thank You.

I'm glad that you found something to help yourself, and I hope you continue to feel better.