Wkat data is accurate

[Andrew1800]

Can someone please advise me if the data on my Lowenstein 20a (trial machine) is correct or if the data I get from Oscar 1.5.3 is.
I have nticed that each morning the machines little screen says AHI 0 and Leak 0 or 2.5 usually. But when I look at the data on Oscar it says AHI of different amounts the highest is AHI 1.53 and the leak rate varies but has been as high as 4 (median).
Does Oscar re analyse the data? this would explain the differences.
Also I have noticed that Oscar says " Your Lowenstein CPAP Device (Model *) has not been tested yet. IT seems sinmilar enough to other devices that it might work but developers would like a .zip copy of this device's SD card and matching clinician .pdf reports to make sure it works with OSCAR"
Would this possibly explain the differenced in results I am getting between the machine and OSCAR? 
Thanks in advance for your help.

[G. Szabo]

I am not familiar with  Lowenstein. But I learned on this forum about the differences between Oscar's and ResMed figures: Oscar reports the data the instrument delivers about the night you select, while the ResMed instrument displays one-month averages unless you change its settings. Hence, there should be an apparent difference.  

[WisNaeMe]

I use a Lowenstein Prisma SmartMAX and can confirm that the numbers displayed on the machine’s ‘little screen’ (or the Prisma phone app for that matter) do not always correspond to the figures displayed when data is imported into OSCAR, so the problem is probably not down to your specific loan machine. 

I’ve always put it down to how values are being rounded, sampled or averaged, so have not been too concerned and as I tend to look more at day-to-day changes rather than the actual values, the numbers have always appeared ‘close enough’ for me. 

These rounded values are maybe more to mirror the smiley faces used by Resmed as indictors on their machines and app.  I think the values you see in OSCAR are the more true representation of your nights sleep.

Others here are importing data into OSCAR from the Lowenstein Prisma 20a and have not reported any concerns about this, but I suspect that this is happening across all Prisma and PrismaLine machines.

You might want to download Lowenstein’s clinical software - PrismaTS, via Apnea Board.

https://www.apneaboard.com/forums/Forum-...-and-Links

It would give you an indication of what Lowenstein regards as the ‘true’ values, but again these don’t match up specifically with data imported into OSCAR (but they are closer than the screen or app).  

It might be that PrismaTS is post-processing some of the card data, whereas OSCAR is only displaying what’s on the card; or what it can (so far) make of the card data.  Lowenstein machines have only recently been supported by OSCAR and so their workings have maybe not yet been fully unravelled …and they do seem to have their own, very particular philosophy on the management and treatment of sleep disordered breathing!

Catching up with apnea board member MACKA might help you too - he writes/edits the Lowenstein WIKI entry.  He’s been chasing down some very specific Lowenstein data dissimilarities/discrepancies, so might have some ideas on this issue.

[Macka]

Hi Andrew. I have same questions. I have been using different model Prisma's and getting different results in Oscar. My latest is a Prisma 30ST and I just noticed it is reporting Tidal volume 100ml less than Oscar.

The Prisma software itself reponds very differently to each model in the range. I think each machine produces the data relevant to the type of therapy and respiratory problem each model is designed to address. I am guessing this means each models data interpretation would likely need to be introduced into Oscars programming also. Would like to know more about how this works.

[Andrew1800]

Hi Andrew. I have same questions. I have been using different model Prisma's and getting different results in Oscar. My latest is a Prisma 30ST and I just noticed it is reporting Tidal volume 100ml less than Oscar.

The Prisma software itself reponds very differently to each model in the range. I think each machine produces the data relevant to the type of therapy and respiratory problem each model is designed to address. I am guessing this means each models data interpretation would likely need to be introduced into Oscars programming also. Would like to know more about how this works.

I have ended up using mainly Prisma TS software, I gather being made by the manufacturer it should be the most accurate. 
Still learning how to read and understand the info correctly.