I'm getting frustrated by this topic because I think we are working towards different goals.
IMHO it has got to be much easier to create a machine to replicate breathing patterns AND responses to algorithmic therapy than it would be to create a crash dummy that reacts a a human being would in an auto crash.
Am I at least right on this one? Come on guys...throw me a bone here!
Finally slept through the whole night....
-
wading thru the muck!
- Posts: 2799
- Joined: Tue Oct 19, 2004 11:42 am
-
rested gal
- Posts: 12881
- Joined: Thu Sep 09, 2004 10:14 pm
- Location: Tennessee
Complicated stuff. In my non-techie opinion, no matter how perfectly one person's sleep disordered breathing pattern were recorded and used for playback in a "replication" machine later, I can't see how that would allow any autopap machine to show it how would perform with the live person in a live sleeping situation.
The breathing feedback from the patient is an absolutely crucial part of making an autopap decide "what to do next". No matter how complex and perfect the recording of a patient's sleep disordered breathing one night was, I don't see how any useful conclusion could be drawn from playing just that recording back in a replication machine and having an autopap supposedly deal with it....without allowing that autopap to get real time feedback from the real patient.
Using only a replication of the patient's sleep disordered breathing, one machine could look like it handled the patterns well and another machine could look like it failed to take correct actions. But the fact would remain... in such a replication test, no matter how perfectly the original recording copied a single person's sleep disordered breathing patterns, during the test itself there'd be no patient breathing back into the machine....no feedback at all for the machine to base its next decision ...its next "moves" on.
A machine that looked "on paper" like it failed (with no feedback breathing) could actually be the best machine of all for that patient if hooked to the same living, breathing patient - this time actually getting feedback for its decisions. Depending on whether a machine was getting real feedback or not could make all the difference in the world, imho, in how it would behave.
Without actual minute to minute feedback breathing from a live patient hooked to it, I don't see how any autopap could show what it truly would do. Replicated breathing patterns from a patient just couldn't give that essential (and often subtle) feedback, imho.
Wader, it really is an intriguing idea...the kind of machine you're talking about. I just don't see how a patient's response could ever be programmed into it other than so crudely that it still wouldn't give meaningful data. I may be all wrong, but each response (patient or machine) depends on the previous response. How would you ever get all those building-upon-each-other responses sorted out?
The breathing feedback from the patient is an absolutely crucial part of making an autopap decide "what to do next". No matter how complex and perfect the recording of a patient's sleep disordered breathing one night was, I don't see how any useful conclusion could be drawn from playing just that recording back in a replication machine and having an autopap supposedly deal with it....without allowing that autopap to get real time feedback from the real patient.
Using only a replication of the patient's sleep disordered breathing, one machine could look like it handled the patterns well and another machine could look like it failed to take correct actions. But the fact would remain... in such a replication test, no matter how perfectly the original recording copied a single person's sleep disordered breathing patterns, during the test itself there'd be no patient breathing back into the machine....no feedback at all for the machine to base its next decision ...its next "moves" on.
A machine that looked "on paper" like it failed (with no feedback breathing) could actually be the best machine of all for that patient if hooked to the same living, breathing patient - this time actually getting feedback for its decisions. Depending on whether a machine was getting real feedback or not could make all the difference in the world, imho, in how it would behave.
Without actual minute to minute feedback breathing from a live patient hooked to it, I don't see how any autopap could show what it truly would do. Replicated breathing patterns from a patient just couldn't give that essential (and often subtle) feedback, imho.
Wader, it really is an intriguing idea...the kind of machine you're talking about. I just don't see how a patient's response could ever be programmed into it other than so crudely that it still wouldn't give meaningful data. I may be all wrong, but each response (patient or machine) depends on the previous response. How would you ever get all those building-upon-each-other responses sorted out?
Last edited by rested gal on Sun Mar 06, 2005 3:45 pm, edited 1 time in total.
Yes, you are right, except... IMO, the intent of the crash test dummy is to measure the GROSS forces experienced by a body of a set shape, weight and charactaristics in an accident, not the subtle ones. So I don't think there's particularly that much lost in translation when we don't account for a fear-induced tightening of muscles, or a "hands thrown in front of face" reaction, because I don't believe that's what they're trying to test.wading thru the muck! wrote:IMHO it has got to be much easier to create a machine to replicate breathing patterns AND responses to algorithmic therapy than it would be to create a crash dummy that reacts a a human being would in an auto crash.
Am I at least right on this one? Come on guys...throw me a bone here!
So yes, I think you are arguing down a different path than I perceived you to be.
S'okay. We still love you.
Liam, coughing and trying to look manly after the uncharacteristic and unmanly show of affection for another man.
_________________
| Machine | Mask | |
 |  |
Sorry for the frustration Wader. Perhaps I just don't understand the concept of a crash test dummy that breaths into an AutoPAP. My point is a breath introduced into an AutoPAP that does not respond in any way whatsovever, necessarily fails to provide any feedback to the machine that is essential to the algorithm. That one-way crash-test-dummy approach necessarily side-steps and potentially breaks the very algorithm that you are trying to test.
I am sure that once again I am not explaining myself adequately. Ugh!
I am sure that once again I am not explaining myself adequately. Ugh!
Yup, Wader! I'm having a hard time reading today without the words jumping around on the page---my own little attention deficit symptom. On days like today I can still analyze a-okay, I can slowly and sequentially type text out just fine (with plenty of rereading and editing), but I cannot easily or accurately read pages of text without my little cognitive problem hindering me. No wonder you are frustrated trying to communicate with me!
However, you are right! The objective would be to create the initial breathing pattern AND the breathing response----exactly as you said!!! Again, I apologize that I am not reading individual words and sentences very well on a day like today. Who knows what other key points you and others have posted that I have missed? Most days are not cognitively as bad as today for me (regarding reading), but I have plenty of days like this. I used to get this problem as a kid too. Untreated sleep apnea just aggravated the heck out of it. With CPAP/AutoPAP I'm back to moderately flawed cognitive life as it was before untreated apnea.
Again, sorry!
However, you are right! The objective would be to create the initial breathing pattern AND the breathing response----exactly as you said!!! Again, I apologize that I am not reading individual words and sentences very well on a day like today. Who knows what other key points you and others have posted that I have missed? Most days are not cognitively as bad as today for me (regarding reading), but I have plenty of days like this. I used to get this problem as a kid too. Untreated sleep apnea just aggravated the heck out of it. With CPAP/AutoPAP I'm back to moderately flawed cognitive life as it was before untreated apnea.
Again, sorry!
-
Guest
Hi Rested Gal,
You are right to the dot....there is no point for a carach dumb simulation except for proto typr test purposues. Nothing beat the real test with full swing real life test. Seem like all the dump test is as good as ready to crash in the market as part of sell first, get feedback, fix some problems, get more new version out and test againand see teh feedback...if works sell more if not good luck to you all that just bought it!
Mckooi
You are right to the dot....there is no point for a carach dumb simulation except for proto typr test purposues. Nothing beat the real test with full swing real life test. Seem like all the dump test is as good as ready to crash in the market as part of sell first, get feedback, fix some problems, get more new version out and test againand see teh feedback...if works sell more if not good luck to you all that just bought it!
Mckooi