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

A question about calculation using OSCAR-reported "Med." values as variables
#1
A question about calculation using OSCAR-reported "Med." values as variables
See   http://www.apneaboard.com/forums/Thread-...#pid442145

Caution:  Here is what may be wild speculation or confusion in a couple of areas. 

This post is about interpreting and discerning differing flow limitation effects and about doing any "Med." value computation based on two or more OSCAR-reported statistical summary values. The intent is to get input on what I'll call a "fuzzy-'data'-analysis" problem. It is not to directly address and confuse member "temperance"'s therapy question with my speculations in his thread linked above. I will post reference to this post in his thread which raises a qood therapy question.

 "temperance" reports low AHI with frequent awakenings and raises apt questions whether his leaks may explain his awakenings. As noted there, leaks alone could cause his awakenings. But IMO his inspiratory time is disproportionately long, causing extra work in breathing and may be adding to or causing his awakenings, if not more leaks.

I haven't replied to him, at least not yet, because I do not fully understand allowable uses of and interrelatedness among OSCAR summary statistics, say, a "Med." respiratory rate, a "Med."  inspiratory time and a "Med." expiratory time as are in his OSCAR Summary report. Again, the the main question I raise is about calculations using fuzzy statistical values and it has come to mind many times. I have not seen it addressed anywhere, but I have wanted to do simple calculations using median values for one or more of respiratory rate, inspiration time or expiration time as below.

IMO, temperance could have a significant and nearly constant flow limitation from, say, hard-tissue upper airway obstruction (unnoted) together with irregularly spaced periods of low level, superimposed soft-tissue (floppy) flow limitation--the latter reflected in "FL" flags. Overall real flow limitation effects, flagged and unflagged, could be additive. Using Resmed's FL scaling for an unflagged hard-tissue flow limitation equivalent to FL= 0.50, for example, his unflagged "FL" would be 0.50. Adding to 0.50 his 99.5% FL up to 0.11, his total "FL" would be 0.61 or higher. Prolonged, that level or near it might be a critical level causing awakening, but that is not the issue here.

Further, hard tissue obstruction might cause some arousals alone and his added soft tissue flow limitation effect could add to it and be positional. 

Illustration of calculation using temperance's median values:

IMO, the OSCAR reported median "Insp. Time" to median "Exp. Time" (I/E=1.94/1.8) ratio would be 1.08. His (inspiratory) duty cycle ratio would be 0.52.  I/(I+E) = 1.94/(1.94+1.8) = 0.52. Those median values, from my computational point of view, are all fuzzy numbers, numbers that simply divide the upper half from the lower half of all numbers in size-sorted number lists. The meaning of ratios based on them is in question. 

Both ratios are a bit high. One research report, using a different overall setup and equipment, reported a duty cycle of about 0.53 as indication of severe upper airway obstruction.  

Here I plug his median Resp. Rate and Insp. Time into a respiratory rate (RR)  formula and solve for "median" Exp. Time  : median Resp. Rate = 15.80= 60 sec/(1.94 + Exp. Time). The result is "median" Exp. Time = 1.85 seconds. The OSCAR summary "Med. Exp. Time" is nearly equal at 1.80 seconds. This single case suggests calculation with medians may be OK on a limited basis.

Please, set me straight if you see this as nonsense.
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.  

Of my 3 once-needed, helpful, and adjunctive devices I have listed, only the accelerometer remains operative (but now idle). My second CMS50I died, too, of old age and the so-so Dreem 2 needs head-positioning band repair--if, indeed, Dreem even supports use of it now.



 
Post Reply Post Reply
#2
RE: A question about calculation using OSCAR-reported "Med." values as variables
Replying to follow
Post Reply Post Reply
#3
RE: A question about calculation using OSCAR-reported "Med." values as variables
If I understand your post, the Insp and Exp median times would need to occur at the same period in time, but I strongly believe that each occurred at a different time interval.  This will skew your data results.  An occlusion of the airway during either breathing phase direction could be marked as a breath.  Based off the definition of "Median", this could skew the data.  Have you tried using the average?  This could possibility smooth the data.  I believe that to really achieve your goal, you would need to calculate the delta of each breath cycle.  As I envision this analysis, you would need to insert logic that includes any breath restriction to be included within that breath cycle direction, until an actual reversal is determined.

DISCLAIMER: I'm only on my second cup of coffee.  Big Grin

Good luck!
Crimson Nape
Apnea Board Moderator
www.ApneaBoard.com
___________________________________
Useful Links -or- When All Else Fails:
The Guide to Understanding OSCAR
OSCAR Chart Organization
Attaching Images and Files on Apnea Board
Apnea Helpful Tips

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
#4
RE: A question about calculation using OSCAR-reported "Med." values as variables
Your calculations aren't valid imo because those statistics are not accurate enough.

If you look at the handful of examples he provided you will see a) inhalation is not longer than exhalation as OSCAR reports and b) cardiogenic oscillations are present at times. The cardiogenic oscillations are causing OSCAR to improperly calculate and report inspiration time, expiration time and respiration rate and since all of those numbers are wrong any calculations and interpretations based off them are not valid.

In short this level of analysis is not possible until OSCAR starts reporting these values accurately.
Post Reply Post Reply
#5
RE: A question about calculation using OSCAR-reported "Med." values as variables
(04-11-2022, 11:16 AM)Geer1 Wrote: In short this level of analysis is not possible until OSCAR starts reporting these values accurately.

OSCAR is reporting what the CPAP says and as such is reporting accurately.  The CPAP on the other hand . . . . . . . .
Post Reply Post Reply
#6
RE: A question about calculation using OSCAR-reported "Med." values as variables
Gideon, OSCAR reports channels that are provided by the CPAP as-is, but channels (out of RR, TV, MV, Ti and Te) that are not present in CPAP data will be calculated by OSCAR. From what I can see AS10 Autoset data has RR, TV and MV channels but is missing Ti and Te so these will be calculated by OSCAR. I agree with Geer1 that these values are not particularly accurate, particularly if cardiogenic artifacts are present - this is the nature of the code and how it treats zero crossings to determine the end of exhalation which will tend to shorten Te and increase Ti.
Post Reply Post Reply
#7
RE: A question about calculation using OSCAR-reported "Med." values as variables
(04-11-2022, 05:59 PM)Gideon Wrote: OSCAR is reporting what the CPAP says and as such is reporting accurately.  The CPAP on the other hand . . . . . . . .

That is what I thought originally but I have since learned that is not the case. 

Autoset machines do not record this information so they have programmed OSCAR to calculate them. The programming is only accurate for simple breathing waveforms and gets confused when things like cardiogenic oscillations are present, when this happens it almost always over calculates inspiration and under calculates expiration. The vauto on the other hand does record this information and is fairly accurate and doesn't run into this issue (at least not to the same degree). The difference is pretty obvious when you compare the two machines data and it makes it pretty clear why you can't rely on these autoset values for interpretation. 

Here is my data on autoset. Median inspiration 1.80, median expiration 1.84.

   

And the next day on vauto. Median inspiration 1.22, median expiration 3.18. Duty cycle cured by changing my machines reporting ability Wink

   
Post Reply Post Reply
#8
RE: A question about calculation using OSCAR-reported "Med." values as variables
Note: This post, below this note, was drafted in Wordpad and completed without benefit of reading kappa's post I just now see at posting time. I and E times from Resmed machines are truly a knotty and challenging problem. A Resmed patent goes to some length showing how their algorithm must see a sufficient rise in inspiratory flow before deciding actual inspiration has started. Still we see cardio waves cross upward across the zero flow rate axis.

@Crimson Nape, median vs. average: Yes, an average would seem far better, but it is not  conveniently available in the OSCAR summary. The skew "positional" or ordinal median values can reflect is why I am uncomfortable with thought of using them as a scalar while still wishing to lazily press them into ratio service. My concern deepens as I type this because it  now occurs to me that I-time skewed high would most likely mean E-time skewed low, vastly overstating I/E for the bulk of data. Correct?

@Geer1;  Yes, I see, thanks to your comment, "I-time < E-time" in the four closer up views (8 min. to 24 min.) of temperance's  flow rate curve strongly suggests those OSCAR-summary times are mostly, if not vastly, wrong for the session. Accordingly, my comments about temperance's disproportionately long inspiration time were wrong.

@Gideon: Data from the Airsense 10 Autoset BRP and PLD files, as far as I know, do not present I and E times so I've assumed OSCAR calculates them (more accurately than it does).  Again, as far as I know, the closest Autoset data related to (the lumped) I and E metric are obscured in the respiratory rate. 

@Anyone: But, on the I and E time topic and for the VAuto data, what is the definition of "B5ITime.2s" and of "B5ETime.2s". Does the "B5" mean 5-breath cycles of data at the 2 second sample rate?

@All three above:  Thank you for correcting me and for helping me see details and the lay of PAP land better. Working with PAP data in Excel spreadsheets without using VBA  is extremely time intensive.  Along the way, I found it best to integrate the flow rate curve first to arrive at the tidal volume of each I-wave including the cardio artifacts. Then select all the TV greater than, say, 25 ml. (I forget the exact choice). With those filtered TV at hand, then use their I and E times for  determination of I and E times. 

One of  the  benefits from comments and commenters cited above: Only as of now, this novice's offers of "help" for finding longer I-times--so as to see durations and how severe they are--must be upgraded. A persistent and telling  wide divergence between an I-time curve (placed above) an E-time curve (placed below I-time curve) must be accompanied by individual inspiration waves which appear to be nearly equal to or of longer duration than their individual expiration half cycles.   

A QUESTION THAT COMES TO MIND OR TO WHATEVER  REMAINS OF  IT Oh-jeez

What's the point, why do I keep putzing around with analysis  and flow limitation? In the large it is my sense and hope I or my reader-betters might find,  develop or stumble upon additional practical self-help; help for many frustrated, poorly-treated, tired, low-AHI  flow limitation sufferers; help for those sufferers who are casting about looking for help at ApneaBoard and elsewhere. There is need.  I've  seen it in calls for help at AB and in my own early experience of PAP-treated high flow limitation level, low AHI, fair sleep care, good pulmonologists, and good GP doctors.
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.  

Of my 3 once-needed, helpful, and adjunctive devices I have listed, only the accelerometer remains operative (but now idle). My second CMS50I died, too, of old age and the so-so Dreem 2 needs head-positioning band repair--if, indeed, Dreem even supports use of it now.



 
Post Reply Post Reply
#9
RE: A question about calculation using OSCAR-reported "Med." values as variables
Kappa, I had previously proposed an idea to pholynyk about an idea to try and improve this calculation but not sure if it ever got considered or if it is possible with the programming capability.

My idea was to try and pinpoint inhalations by targeting a high positive flow rate (5, 10 or whatever seems reasonable) followed by a drop to at least 0 (my first idea was a negative flow rate which might work but 0 might be more accurate). Respiration rate then calculated as number of inspirations/time. Inspiration time is calculated as total positive flow rate time/number of inspirations and expiration time is the remainder or total negative flow rate/number of inspirations.

My idea was to try and get away from relying on when flow rate crosses zero. Flow rate can cross zero too often to be a reliable measure although most of the time it crosses zero improperly it will be with a small amplitude hence my idea to target high enough flow rates to be considered an obvious inhalation. Imo the odd times that there are large amplitude fluctuations over zero (rem breathing, arousal breathing, the cathyf type fluctuating breaths etc) can be improperly recorded as they are less important and relatively rare.
Post Reply Post Reply


Possibly Related Threads...
Thread Author Replies Views Last Post
  Interpreting Oscar values SleepingTight 0 139 03-10-2024, 06:53 PM
Last Post: SleepingTight
  OSCAR Graph Question Events Flags/AHI Bob Cali 3 477 09-28-2023, 02:53 PM
Last Post: Gideon
  [Equipment] OSCAR Question: AHI Score from Wellue ring not showing phoenixsaga 10 1,167 09-12-2023, 11:21 AM
Last Post: Crimson Nape
  Descriptions of variables measured by OSCAR BlueSky1958 1 364 07-24-2023, 10:57 AM
Last Post: Crimson Nape
  Question about OSCAR accuracy during high leak periods raptosnorus 13 921 06-26-2023, 12:55 PM
Last Post: raptosnorus
  40 second OA reported in OSCAR - can that be right? shenstone 2 582 03-18-2023, 08:18 PM
Last Post: Sleeprider
  Need help reading reported CAs as "true" or "false" JoyD. 1 485 01-25-2023, 04:07 PM
Last Post: quiescence at last


New Posts   Today's Posts


About Apnea Board

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