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

Just don'tunderstand ?
#11
RE: Just don'tunderstand ?
(05-01-2015, 09:56 AM)DariaVader Wrote: not sleeping also lowers it.... js


Agree!
So the low AHI isn't something to get excited over. I'd rather feel more rested.

OpalRose
Apnea Board Administrator
www.apneaboard.com

_______________________
OSCAR Chart Organization
How to Attach Images and Files.
OSCAR - The Guide
Soft Cervical Collar
Optimizing therapy
OSCAR supported machines
Mask Primer



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
#12
RE: Just don'tunderstand ?
We can't assume that the algorithms that judge exactly what is an apnea hypopnea, are infallible, either. Especially if your treated AHI is low this means that many of your events are on the edges of the range defined for them, with a consequent rise in likely errors. I frankly doubt that there is actually resolution down to three significant places but the software reports to that level. All sorts of external events could also affect the measurements. Earthquakes, moving around in bed, heck even loud noises could affect the measurement accuracy.

The point is not to obsess over the last decimal point. I suspect myself that an accuracy for the AHI to +- 0.5 is really about all we should expect.
Ed Seedhouse
VA7SDH

Part cow since February 2018.

Trust your mind less and your brain more.


Post Reply Post Reply
#13
RE: Just don'tunderstand ?
(04-30-2015, 08:57 PM)PaytonA Wrote:
(04-30-2015, 08:31 PM)Mr. Van Winkle Wrote: How you can go from a .04 AHI to a 3.2 AHI with out knowing your doing anything different?
Just don't get it, I feel the same just differnt numbers.
????
Sleep-well

Are you sure that was not a 0.4 AHI night? In order to score a 0.04 AHI one would have to sleep for 25 hours and have 1 event.

Yes, My bad .4 AHI , But I did hit 0 once.
Anyway I'm doing great with this CPAP gig , I'm just going to not let myself get caught up with the numbers when I feel great!
37 was my study number as long as I stay under 5 I'm good...
I still need to peek at em every day I just need to not worry about it.
Heck I have some friends on CPAP that never look at their numbers.

Sleep-well
Post Reply Post Reply
#14
RE: Just don'tunderstand ?
(05-01-2015, 12:34 PM)eseedhouse Wrote: We can't assume that the algorithms that judge exactly what is an apnea hypopnea, are infallible, either. Especially if your treated AHI is low this means that many of your events are on the edges of the range defined for them, with a consequent rise in likely errors.

I do not understand what you are saying here. Scoring an event is an all or nothing proposition. It either meets the requirements or it does not.

(05-01-2015, 12:34 PM)eseedhouse Wrote: I frankly doubt that there is actually resolution down to three significant places but the software reports to that level. All sorts of external events could also affect the measurements. Earthquakes, moving around in bed, heck even loud noises could affect the measurement accuracy.

Earthquakes, moving around in bed, and loud noises might be scored as snore but snores do not count in AHI. if any of those things caused a person to hold their breath for over 10 seconds then, guess what, it is an apnea. If the person was awakened prior to the event, then it obviously should be ignored but the machine scored it correctly.

(05-01-2015, 12:34 PM)eseedhouse Wrote: The point is not to obsess over the last decimal point.

I agree with this wholeheartedly.

(05-01-2015, 12:34 PM)eseedhouse Wrote: I suspect myself that an accuracy for the AHI to +- 0.5 is really about all we should expect.

I am quite sure that the machine can total the events very accurately. I also think that the machine can total the time quite accurately. Since those are the only two numbers that go into the AHI calculation, I think that the accuracy is much better than +-0.5.

The major problem that I see is that the machine can not tell reliably when you are actually asleep and that is where human judgement must come in to play and if that is what your concern is with the accuracy of the machine reported AHI, I must agree.

Best Regards,

PaytonA

Admin Note:
PaytonA passed away in September 2017
Click HERE to read his Memorial Thread

~ Rest in Peace ~
Post Reply Post Reply
#15
RE: Just don'tunderstand ?
Is there anything we can do to make our sleep better besides using our cpap every night?

I have diabetes, type 1, and when I look at the numbers from my blood glucose monitor - I know durn well why I get a higher number than I should be seeing. With the cpap machine, it is all a mystery. From reading the posts on this board, I am assuming that one should not snack before you go to bed. What else do you do to insure a good night's sleep?
Post Reply Post Reply
#16
RE: Just don'tunderstand ?
It isn't that the machine can't count, it's that the definition of just what is and what isn't an "event" is and must by the nature of things be imprecise. Breathing is an irregular process and no instrument is ever going to be able to measure it perfectly. So we say that an "event" has to last 10 seconds to be counted. Well is an event that's 9.9999 seconds long any better for you than one that's 10.00001 seconds long? Yet the first is counted and the second is not. The real world is analogue, not digital. There are *always* measurement errors, there *must* be measuring errors because that's the way our universe works.

It's well known in all the sciences that you must determine your error limits for *any* measurement. False precision is one of the bigger traps in science.

I'm not complaining - you have to draw the line somewhere after all - but the real world is not cut up into nice even chunks, it's wiggly. Each breath fades into the next and there's no sharp line between them. Every machine brand has a slightly different algorithm.
Ed Seedhouse
VA7SDH

Part cow since February 2018.

Trust your mind less and your brain more.


Post Reply Post Reply
#17
RE: Just don'tunderstand ?
(05-01-2015, 04:19 PM)eseedhouse Wrote: It isn't that the machine can't count, it's that the definition of just what is and what isn't an "event" is and must by the nature of things be imprecise. Breathing is an irregular process and no instrument is ever going to be able to measure it perfectly. So we say that an "event" has to last 10 seconds to be counted. Well is an event that's 9.9999 seconds long any better for you than one that's 10.00001 seconds long? Yet the first is counted and the second is not. The real world is analogue, not digital. There are *always* measurement errors, there *must* be measuring errors because that's the way our universe works.

It's well known in all the sciences that you must determine your error limits for *any* measurement. False precision is one of the bigger traps in science.

I'm not complaining - you have to draw the line somewhere after all - but the real world is not cut up into nice even chunks, it's wiggly. Each breath fades into the next and there's no sharp line between them. Every machine brand has a slightly different algorithm.

Actually,, I agree with you for the most part. Just like you said, every measurement has error. Like you also said does it matter to us that it was 10.00001 instead of 9.9999.That would be an error. Is it important in our frame of reference? No it is not.

The machine merely reports events that fit the *very precise definition for scoring an apnea or hypopnea*. Does it do this accurately? Yes it does it very accurately. Is there any error in those measurements? Yes there is but it would be somewhere down in the 5th or 6th decimal place. For our purposes the errors are rather trivial. There must be judgement involved with interpreting the data provided by the PAP. For instance, I think that most of us are aware that apneas of 10-20 seconds are nothing to worry about unless they are in clusters. The machine does not tell us how large a cluster needs to be to be important. We must make that decision. We also know that a single 10 second apnea is not at all serious so the fact that it is 10.00001 instead of 9.9999 seconds duration really does not mean much.

I think that the reported AHI is quite accurate but as we have all said, it is only part of the story and it needs human judgement to properly assess its impact. We sometimes, myself included, get all wrapped up in the numbers and forget that they are just tools.

"The real world is analogue not digital". I agree but we digitize it in order to attempt to understand it.

Best Regards,

PaytonA

Admin Note:
PaytonA passed away in September 2017
Click HERE to read his Memorial Thread

~ Rest in Peace ~
Post Reply Post Reply
#18
RE: Just don'tunderstand ?
(05-01-2015, 09:31 PM)PaytonA Wrote: The machine merely reports events that fit the *very precise definition for scoring an apnea or hypopnea*.

But does every machine have the same "very precise" definition?
I would expect they don't. Do you have any actual evidence to the contrary?

Quote:Is there any error in those measurements? Yes there is but it would be somewhere down in the 5th or 6th decimal place.

Well, for instance last night my machine scored 0.6 "events" per hour. Sleepyhead for the same night from the same data gives AHI as 0.7 - a long way from the 5th decimal place. Utterly irrelevant of course, but not the fifth decimal place.

Since we can't really compare two machines on the same patient for the same time we have no way to show that any two machines will give the same AHI for the same night and patient. And of course the machine itself alters the AHI, since that is of course it's very purpose. And worse, every brand alters things differently with proprietory algorithms that they all claim are better than the competition's. That makes cross brand comparisons pretty hard to quantify but we would certainly expect them to exist.

Quote:For our purposes the errors are rather trivial.

I agree - but I'd say that the significance of a difference in AHI of 1 or two events per hour is also trivial.

Quote:"The real world is analogue not digital". I agree but we digitize it in order to attempt to understand it.

Digitization is a useful tool. Essential to understanding the world? I rather doubt that. Science made big strides using purely analogue measurements long before digital was even invented.

And I don't think any scientists claims that we really understand the world. Significant parts of it, sure. But you can't even see the whole universe, let alone "understand" it in the intellectual sense, because you can never stand outside it.


Ed Seedhouse
VA7SDH

Part cow since February 2018.

Trust your mind less and your brain more.


Post Reply Post Reply
#19
RE: Just don'tunderstand ?
(05-01-2015, 10:15 PM)eseedhouse Wrote: But does every machine have the same "very precise" definition?
I would expect they don't. Do you have any actual evidence to the contrary?

I do not know this for a fact but I would imagine that the manufacturers use the "Rules For Scoring Respiratory Events In Sleep" in the "AASM Manual For Scoring of Sleep and Associated Events" which is very precise.

(05-01-2015, 10:15 PM)eseedhouse Wrote: Well, for instance last night my machine scored 0.6 "events" per hour. Sleepyhead for the same night from the same data gives AHI as 0.7 - a long way from the 5th decimal place. Utterly irrelevant of course, but not the fifth decimal place.

One rounds and the other truncates?????


(05-01-2015, 10:15 PM)eseedhouse Wrote: Since we can't really compare two machines on the same patient for the same time we have no way to show that any two machines will give the same AHI for the same night and patient. And of course the machine itself alters the AHI, since that is of course it's very purpose. And worse, every brand alters things differently with proprietory algorithms that they all claim are better than the competition's. That makes cross brand comparisons pretty hard to quantify but we would certainly expect them to exist.

The original question was why the difference in AHI from one night to the next with the same machine.

(05-01-2015, 10:15 PM)eseedhouse Wrote: I agree - but I'd say that the significance of a difference in AHI of 1 or two events per hour is also trivial.

Agree.

(05-01-2015, 09:31 PM)PaytonA Wrote: "The real world is analogue not digital". I agree but we digitize it in order to attempt to understand it.

(05-01-2015, 10:15 PM)eseedhouse Wrote: Digitization is a useful tool. Essential to understanding the world? I rather doubt that. Science made big strides using purely analogue measurements long before digital was even invented.

And I don't think any scientists claims that we really understand the world. Significant parts of it, sure. But you can't even see the whole universe, let alone "understand" it in the intellectual sense, because you can never stand outside it.

We use things like clocks and rulers to help us understand the world. These are digitizations of analogues. Digital does not simply apply to semiconductors.

Best Regards,

PaytonA

Admin Note:
PaytonA passed away in September 2017
Click HERE to read his Memorial Thread

~ Rest in Peace ~
Post Reply Post Reply
#20
RE: Just don'tunderstand ?
I slept 8 hours last night. I had one event, a CA in the last hour! I had AHI of 0.12. The CA lasted 3600 seconds. I love being under 0.2 AHI, but find I cannot express this. The others in the room are not as happy. They are talking about me - say I stopped breathing.

Probably if I had a look at the next significant digit, it would tell me a lot more precisely my condition and quality of sleep.
Dedicated to QALity sleep.
Post Reply Post Reply



New Posts   Today's Posts


About Apnea Board

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