Apnea Board Forum - CPAP | Sleep Apnea
Comparing ResScan to Sleepyhead over a long period - Printable Version

+- Apnea Board Forum - CPAP | Sleep Apnea (https://www.apneaboard.com/forums)
+-- Forum: Public Area (https://www.apneaboard.com/forums/Forum-Public-Area)
+--- Forum: Software Support Forum (https://www.apneaboard.com/forums/Forum-Software-Support-Forum)
+--- Thread: Comparing ResScan to Sleepyhead over a long period (/Thread-Comparing-ResScan-to-Sleepyhead-over-a-long-period)



Comparing ResScan to Sleepyhead over a long period - calcuttaman - 11-22-2017

I haven't compared the two software's in a long time and I just want to understand the differences.

It looks like Sleepyhead does not include the days not used in there calculations and ResScan does.


RE: Comparing ResScan to Sleepyhead over a long period - pholynyk - 11-22-2017

You haven't shown us the number of events recorded by ResScan, so I can't determine how ResScan calculated the AHI, so I can't say if ResScan actually used the unused days in calculating the average AHI.

They shouldn't, because the number of events on those days is unknown, as are the number of hours of sleep. Thus we have an unknown number of events, and adding that unknown event count into all the known event counts gives an unknown total to be divided by an unknown total hours.

The same is true for the SleepyHead data, although I have my own data, and the code, so I can (eventually) figure out how SleepyHead makes it's calculations (which may be incorrect). For example, on the Overview page, it appears that the 'Normal Average' of the AHIs is calculated by adding up the daily AHIs and dividing by the number of days - but it is not clear to me that the average of a set of averages is really the sum of the set divided by the count of the set. That's a question for another day with fewer other tasks to work on.


RE: Comparing ResScan to Sleepyhead over a long period - calcuttaman - 11-22-2017

(11-22-2017, 11:24 AM)pholynyk Wrote: You haven't shown us the number of events recorded by ResScan, so I can't determine how ResScan calculated the AHI, so I can't say if ResScan actually used the unused days in calculating the average AHI.

They shouldn't, because the number of events on those days is unknown, as are the number of hours of sleep. Thus we have an unknown number of events, and adding that unknown event count into all the known event counts gives an unknown total to be divided by an unknown total hours.

The same is true for the SleepyHead data, although I have my own data, and the code, so I can (eventually) figure out how SleepyHead makes it's calculations (which may be incorrect). For example, on the Overview page, it appears that the 'Normal Average' of the AHIs is calculated by adding up the daily AHIs and dividing by the number of days - but it is not clear to me that the average of a set of averages is really the sum of the set divided by the count of the set. That's a question for another day with fewer other tasks to work on.
As far as I can tell I can't find the total number of apneas over a given date range in either ResScan or Sleepyhead.  Do you know of a way?


RE: Comparing ResScan to Sleepyhead over a long period - pholynyk - 11-22-2017

Not offhand, although it might be in the exported csv - I'll have to check. Of course it shows up for each day on the Events tab of the daily page. One could take a short range - like a week - and add things manually and then compare with the Overview or Statistics figures.

ETA - the various event type counts do show up in the csv files, both the session and summary files.