Respiratory Rate wrong while abnormal breathing?

I was just about to make a post about my weird high respiratory rate. But I watched that freecpapadvice.com guy yesterday (he's certified and has a lot to share on his youtube). His videos made me think: "sometimes the machine is wrong at categorizing".

My chart shows ~50 RR. But I count the actual breaths in a minute (during my slightly irregular breathing time) and it's 25. (So I'm not hyperventilating here! )

"Interesting" and I hope one day SleepyHead can do some second-guessing of the machine's data. (I come from the developer world, so I know that's possible, at least in theory.) Is SleepyHead the right place to wait for features like this, or are there greener pastures elsewhere?

The software available....either SleepyHead or ResScan really only are reporting the data that the machine gathers.
So the machine is gathering that double dipping respiration rate and the software simply reports it.
What you are seeing is quite common and discussed fairly often here. The machine is picking up a cardiac echo and counting it as a respiration. Obviously it isn't.

So it's not so much the software reporting or evaluating but it all starts with what the machine senses and records.
It's not so much modifying the software that is needed to adjust for the cardiac echo as it is modifying the machine sensing itself so that the machine doesn't record double respirations caused by the cardiac echo effect.
While we do see this fairly often...it still isn't common and the people who get the double respirations recorded are in a minority.
Most people won't be seeing it.

Pugsy wrote: ↑

Wed Apr 17, 2019 9:55 am

The software available....either SleepyHead or ResScan really only are reporting the data that the machine gathers.
So the machine is gathering that double dipping respiration rate and the software simply reports it.

Wow. So... bad firmware on the machine. Hah. Kind of funny to me that it's entirely closed source and you'd have "no chance" of fixing it. (Without working for Resmed themselves? Opening an issue on their github page? Buying an extra machine and whipping out the 'ol JTAG debugger?)
I'm sure it could be fixed by SleepyHead after the fact, but I agree that the machine itself should be correctly interpreting this.

I always forget the big long official term

Ballistocardiographic artifacts

viewtopic.php?f=1&t=174268&p=1283319&hi ... o#p1283319

viewtopic/t61466/Central-Apneas-and-Res ... 45#p576919

You have to understand that the respiration data point gathered is rarely used by anyone who ResMed tries to make happy with machine functions. ResMed looks at the DMEs and doctors as their "customers" not us peon users. The medical profession doesn't really monitor respiration rate with the data from a cpap/apap machine. If there is a potential issue with respiration rate there are probably better choices for such monitoring and it really only comes into play when people are needing special ventilation....those higher end cpaps targeted for special needs.

Add in the fact that you are in a small minority seeing the double counting...no one at ResMed really cares because the doctors or DMEs won't care since you are using a plain ordinary cpap/apap machine and not one of those other higher end machines where respiration rate and volume is more important because of special needs.

You fall through the cracks because it really isn't a critical issue to anyone but you.
Past software versions didn't even report respiration rate from a cpap/apap and I would have to look to see if the current ResScan even reports it on cpap/apap machines (I simply don't remember). The data has always been there but what gets documented by ResScan has often depended on which model machine is being used.

Pugsy wrote: ↑

Wed Apr 17, 2019 9:55 am

The software available....either SleepyHead or ResScan really only are reporting the data that the machine gathers.

Does Resscan count those tiny fluctuations as breaths?

I figured it was just SH not having any sanity checks against bad data.

palerider wrote: ↑

Wed Apr 17, 2019 11:29 am

Does Resscan count those tiny fluctuations as breaths?

I figured it was just SH not having any sanity checks against bad data.

You know I don't really know the answer to that question. I have never had them so I can't compare the rate between the 2 software products. I never looked because I didn't seem to recall even seeing RR on ResScan with my machines over the years.

Sooooo....I just fired up ResScan...and guess what...zero respiration rate data at all on this AirSense 10 machine ResScan software report. The only thing that even offers RR rate is the summary graphs and it's checked but the graph is blank. It's not even offered on the detailed sections.
It's one of the features that the machine does gather the data but because of the model of machine being used it doesn't show the data in ResScan. That's a data point that is really only useful on those higher end fancy machines so ResMed in their ultimate wisdom has ResScan simply ignoring that data point in my lowly AirSense machine. If I remember correctly it was the same with the S9s that I had.

So.....there is no way to know what ResScan says about RR with the AirSense machine to even compare it to what SH says.
A long time ago I read something somewhere that confirmed my suspicions....what the machine can and does record doesn't always show up on the ResScan reports. It all depends on the model of machine that is generating the reports.
There is data there but ResScan doesn't necessarily show it. When ResScan doesn't show it we can't compare it to SleepyHead.

Now you might have RR rate show up on ResScan if you used ResScan because you are using a bilevel machine....If I have the time (which I don't have now) I will try to go back and find some of my old S9 bilevel reports to see if ResScan shows it...but I have other fires to put out at the moment and it's not a critical issue anyway. Not to mention I have never seen any evidence of my having the cardiac echo thing...so even it ResScan showed RR...I wouldn't have any known RR times to compare to.

My question is whether the data is even recorded on the card... It may not be, it may be that SH is just counting all zero crossings and creating it, since it's there for some machines... Like other things that SH calculates... (Like the I:E ratio.) And probably shouldn't considering the confusion it frequently causes.

palerider wrote: ↑

Wed Apr 17, 2019 1:02 pm

My question is whether the data is even recorded on the card... It may not be, it may be that SH is just counting all zero crossings and creating it, since it's there for some machines... Like other things that SH calculates... (Like the I:E ratio.) And probably shouldn't considering the confusion it frequently causes.

You know that's very possible. SleepyHead playing doctor again. I originally thought that the RR was on the SD card because I thought I remembered seeing it on some ResScan reports....and maybe what I was remembering was one of those other higher end machines.
Which was why I said SH just reports what the machine records...but I am old and my memory sucks and I often dream up stuff.

palerider wrote: ↑

Wed Apr 17, 2019 11:29 am

Does Resscan count those tiny fluctuations as breaths?

As Pugsy says, ResMed does not report RR on non ventilator machines, i.e., cpap and auto cpap. It does report RR on those machines considered to have some ventilator capabilities, i.e., Vauto and above.

I do not see any instance where ResScan has been fooled by cardio artifacts on my limited number of Vauto reports. But, my data is very sparse and is limited to one patient, so my data in not conclusive.

I do know that, obviously, the pulmonary people are aware of these artifacts and some machines use techniques to suppress the effects. One technique that is used is to employ a bandwidth filter which removes the cardio wave from the respiratory wave. I don't remember what manufacturer uses this method, maybe more than one. A patent search might yield a clue.

In any event, my bet would be that in machines where the Respiration Rate can be affected (ventilators) some method to remove these artifacts is used.

Maybe Morbius would be kind enough to tell us.

One thing occurs to me. Even though ResScan doesn't report RR for non ventilators, it does report Minute Ventilation which is a function of RR. Therefore, it seems, even though ResScan doesn't report RR it must be being used.

palerider wrote: ↑

Wed Apr 17, 2019 1:02 pm

My question is whether the data is even recorded on the card... It may not be, it may be that SH is just counting all zero crossings and creating it, since it's there for some machines... Like other things that SH calculates... (Like the I:E ratio.) And probably shouldn't considering the confusion it frequently causes.

Data is coming from the card. Been using SleepyHead for years, but first time pouring through the code because this was questioned.

sleepyhead/SleepLib/loader_plugins/resmed_loader.cpp:275

Many references in there to RespRate which it's getting from the card.

Well, there have been other cases of SH not doing ant sanity checking on it's data... And counting every zero crossing as a breath is just another example. When you have 50-60 excursions, counting those tiny 1-2 level pulsations as breaths is just... Wrong.

SH shouldn't even make up data if it's not on the card.

I'll lobby for Oscar to not follow in those footsteps.

linagee wrote: ↑

Wed Apr 17, 2019 2:29 pm

palerider wrote: ↑

Wed Apr 17, 2019 1:02 pm

My question is whether the data is even recorded on the card... It may not be, it may be that SH is just counting all zero crossings and creating it, since it's there for some machines... Like other things that SH calculates... (Like the I:E ratio.) And probably shouldn't considering the confusion it frequently causes.

Data is coming from the card. Been using SleepyHead for years, but first time pouring through the code because this was questioned.

sleepyhead/SleepLib/loader_plugins/resmed_loader.cpp:275

Many references in there to RespRate which it's getting from the card.

It's there on some cards, yes, but is it there on *your* card?

Just wanted to mention that I also have cardiogenic squiggles, but smaller ones. While they don’t affect the SH numbers for my respiration rate, they do distort the measures of my exhalation times. Go figure.

A couple of excerpts from a paper we've seen before:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4629962/
Treatment of sleep-disordered breathing with positive airway pressure devices: technology update

In order to maintain a stable mask pressure, the microprocessor must adjust the turbine speed in response to deviations in pressure that occur from leak or normal swings in air pressure from breathing. The flow signal is sent through low and high pass filters to separate the respiratory flow signal from artifacts. Low pass filters exclude large quick deviations in flow (eg, coughs or sneezes) and high pass filters exclude cardiogenic fluctuations

Determination of the inspiratory and expiratory cycles is essential not only to provide bilevel PAP but also for expiratory pressure relief and determination of inspiratory flow limitation in auto algorithms...
SV and some auto devices use more continuous respiratory cycle determination. ResMed’s servoventilator microprocessor uses fuzzy inference rules looking at the flow rate (relative to mean flow), direction, and size to determine the phase of the respiratory cycle

Which could mean the Respiratory Rate being presented by Sleepyhead is before actions are taken on the Flow Signal to remove artifacts and the applied fuzzy logic determination - or not.