Possible SH bugs???

A poster over on apneaboard posted the following today:

apneaboard poster on apneaboard wrote: Hi, I have a Dreamstation and I am comparing the output data in Sleepyhead and Encorepro2.  I have found that most events tally. However the number of recorded Hypopnea vary drastically. Circa 75% more recorded in EncorePro2. This obviously makes the average AHI different.  Is this a known error ? Thank you.

I've posted this over there, but thought it relevant to post here as well:

In the past, I never noticed any discrepancies between my PR BiPAP data and Encore's. But since getting the DreamStation last fall, I'll be honest, I haven't been looking at the data in Encore very much.

In spot checking my data right now, I have noticed that the last two nights (which Encore downloaded flow rate data for are spot on. But some of the days which Encore did not download flow rate data for do indeed have discrepancies between how the two programs report data. And it's not just the hypopneas. On one day, SH says I had 0 CAs, but Encore said I had 1 CA (a bit after 4:00) for a CAI = 0.2.  Wish I had the ability to see what the Encore flow rate looked like for that day.

I'll try to dig through my own data a bit more this evening, but I might not get around to it. I will post some screenshots when I have a better idea of how often this might be happening.

In the meantime, anybody else notice discrepancies between EncorePro2's view of DreamStation data and SH's?

I use Resmed S9 and often I compare Sleepyhead and ReScan and found them identical...possibly something with the way Dreamstation counts events that Encore picks up but Sleepyhead doesn't.

I don't know about the Dreamstation, but I know if SleepyHead doesn't get the ResMed Detail Data, the count will be off from ResScan. Apparently, Sleepyhead uses the detail data of ResMed for some of the AHI calculations. Otherwise, if Detail Data is present, the counts are nearly identical as LSAT said.
Sounds like the Dreamstation may be similar where discrepancies occur when Flow Rate data is not present.

Jay Aitchsee wrote:I don't know about the Dreamstation, but I know if SleepyHead doesn't get the ResMed Detail Data, the count will be off from ResScan. Apparently, Sleepyhead uses the detail data of ResMed for some of the AHI calculations. Otherwise, if Detail Data is present, the counts are nearly identical as LSAT said.
Sounds like the Dreamstation may be similar where discrepancies occur when Flow Rate data is not present.

In the data that I've checked, SleepyHead downloaded the flow rate data, along with everything else on the days where the data differs with Encore.

The problem is that Encore does NOT download the flow rate data except for the very last night (or two if you're lucky). And so I can't compare the flow rate data that exists in SleepyHead to the flow rate data suppressed by Encore during the Encore downloads.

Back when JediMark was still actively working on SH, I used to do a lot more spot checking SH data vs. Encore data as one of JediMark's testing group. But that was on my old PR System One BiPAP. And once we got into the 0.9-betas, the data in SH and the data in Encore always agreed.

So there may be a bug in the code that was written to make SH work with the DreamStations. I don't code, so I don't know what was done inside of SH to make SH be able to read the DreamStation data.

At any rate, I've now generated Encore detailed data reports for every day since I got the DreamStation. It's going to take some work to try to figure out how often the discrepancies occur and whether there's any kind of pattern in them. I haven't got any hypotheses about this going in. And I will update the forum once I at least know whether the problem affects a small number of days or a significant number of days.