Which is more accurate? Machine or Sleepyhead?

General Discussion on any topic relating to CPAP and/or Sleep Apnea.
User avatar
Turntex
Posts: 43
Joined: Sat Sep 23, 2017 8:24 pm

Which is more accurate? Machine or Sleepyhead?

Post by Turntex » Sun Nov 12, 2017 7:11 pm

Two mornings ago I awoke and looked at my Resmed Aircurve 10 Vauto display and it showed an AHI of 20 for the night's sleep. That is the worst I have ever seen and was really disappointed. I loaded my data into Sleepyhead to see what caused it and Sleepyhead shows an AHI of 3.4 which is very normal for me. Nothing stuck out on the Sleepyhead data that was unusual to cause the machine to read so high. Which is the one to trust? Which is more accurate?

User avatar
TASmart
Posts: 985
Joined: Wed Jan 04, 2017 7:23 pm
Location: Eugene, OR

Re: Which is more accurate? Machine or Sleepyhead?

Post by TASmart » Sun Nov 12, 2017 7:16 pm

You need to go over to apneaboard and get the link to download ResScan and see what it shows. It works like Sleepyhead but as the ResMed software, it will add confidence to your numbers.
All posts reflect my own opinion based on my experience and reading.
Your mileage may vary
Past performance is no guarantee of future results
Consult with your own physician as people very

User avatar
Pugsy
Posts: 43907
Joined: Thu May 14, 2009 9:31 am
Location: Missouri, USA

Re: Which is more accurate? Machine or Sleepyhead?

Post by Pugsy » Sun Nov 12, 2017 7:20 pm

SleepyHead just shows what the machine records. SH doesn't do any flagging or anything like that. SH doesn't do any actual analyzing for the air flow to cause the flags. The machine does it and writes the data to the SD card and then SH just reports it.
When in doubt you can always use ResScan for confirmation.
The other night with the AHI on the screen of 20....had you turned the machine off during that night for some reason or other....what was the number of hours for that reported session showing 20 AHI?
Can you post the detailed report from SH for that night in question?
I may have to RISE but I refuse to SHINE.

User avatar
Turntex
Posts: 43
Joined: Sat Sep 23, 2017 8:24 pm

Re: Which is more accurate? Machine or Sleepyhead?

Post by Turntex » Sun Nov 12, 2017 11:07 pm

I'm not so sure about you guys comments that Sleepyhead does not calculate or interpret data. Almost every night the AHI reported on my machine is different than what Sleepyhead says. It is usually not a lot but is still different. For example, last night my machine reported an AHI of 1.79. Sleepyhead says the AHI is 1.62. It is almost always different some. Not sure what is up with that. I will try to post the data from the night the machine reported the 20 AHI as soon as I get a chance.

Also, I tried installing ResScan but it will not run. I keep getting an error "Pure Virtual Function Called". I did some searches and followed the 2 or three different suggestions I found on the other forum on how to fix this but none of them work! Not a big deal so I just gave up. Have too many other things to mess with than trying to get some stupid software to work. And no, I did not install the drivers and yes, I did uninstall and clean up the registry with CC Cleaner, then reinstall.

User avatar
palerider
Posts: 18503
Joined: Wed Dec 16, 2009 5:43 pm
Location: Dallas(ish).
Contact:

Re: Which is more accurate? Machine or Sleepyhead?

Post by palerider » Sun Nov 12, 2017 11:49 pm

Turntex wrote:I'm not so sure about you guys comments that Sleepyhead does not calculate or interpret data. Almost every night the AHI reported on my machine is different than what Sleepyhead says. It is usually not a lot but is still different. For example, last night my machine reported an AHI of 1.79. Sleepyhead says the AHI is 1.62. It is almost always different some. Not sure what is up with that. I will try to post the data from the night the machine reported the 20 AHI as soon as I get a chance.
Likely slightly different time frames, especially if you're sleeping past machine noon.

_________________
Machine: S9 VPAP™ Auto BiLevel Machine
Mask: AirFit™ P10 Nasal Pillow CPAP Mask with Headgear
Additional Comments: S9 Vpap Auto

User avatar
Turntex
Posts: 43
Joined: Sat Sep 23, 2017 8:24 pm

Re: Which is more accurate? Machine or Sleepyhead?

Post by Turntex » Mon Nov 13, 2017 2:08 am

I am not sleeping past machine time so that should not be it.

User avatar
Jay Aitchsee
Posts: 2677
Joined: Sun May 22, 2011 12:47 pm
Location: Southwest Florida

Re: Which is more accurate? Machine or Sleepyhead?

Post by Jay Aitchsee » Mon Nov 13, 2017 6:40 am

Turntex wrote:I'm not so sure about you guys comments that Sleepyhead does not calculate or interpret data. Almost every night the AHI reported on my machine is different than what Sleepyhead says. It is usually not a lot but is still different. For example, last night my machine reported an AHI of 1.79. Sleepyhead says the AHI is 1.62. It is almost always different some. Not sure what is up with that. I will try to post the data from the night the machine reported the 20 AHI as soon as I get a chance.
SleepyHead does calculate the AHI based on the time line extracted from the detailed data. This time could be slightly different than that used by the machine due to different rounding techniques. If there is no detailed data, the AHI calculated by SleepyHead will be in error. Additionally, there are some SleepyHead settings which could cause the machine and SleepyHead results to differ. Under preferences in SleepyHead, using the "combine close sessions" and/or "ignore short sessions" may cause the machine and SleepyHead time lines to differ slightly, as well as the "Cumulative Indices" Setting which allows several choices in the method of computation of the indices which would produce slightly differing results.

AHI= number of events/number of hours. In the above example, the difference in AHI is 1.79-1.62, or 0.17. Over 8 hours this is the equivalent of about 1 event (1/8=0.13). Further, if we were to work backwards using an assumed 8 hour night, 1.79*8=14.32 events, while 1.69*8=13.52 events. Stated another way, both AHI ratios are equivalent to 14 events over 8 hours since events must be whole numbers. Over eight hours, a difference of 0.17 does not appear to be significant. Further, in my experience, ResScan and the ResMed machine report to the nearest 10th, not hundredth, but I have an S9, the Airsense could be different.

As for your first post in this thread where there was a perceived machine reading of 20 and SleepyHead reported 3 something, I'm guessing something caused an incorrect machine display, but I haven't a clue what that could have been.

I have seen an instance recently where SleepyHead recorded multiple sessions for the same time frame resulting in an AHI computed by SleepyHead that was double that reported by the machine and ResScan (if there was one session for the night that was recorded twice). The fix (for me) was to "Rebuild CPAP Data".