03-14-2022, 12:47 PM
Data structure / data definition of AirSense 11 CD, CMSd50+ or OSCAR?hing obvious, …
Hi, I’m trying to get the time series data into a separate tool for statistical analysis (Mathematica if that helps). Currently using AirSense 11 and CMSD50+ And reading into OSCAR (current 3.1). I’m looking for directory and file structure documentation that would let me read the stored data directly. Since I’m very interested in the O2, I started with the O2meter. The reader produces 2 files. The excel CSV seemed of interest until I saw it didn’t include the internal flags or in any other way (besides their values) distinguish between valid and bogus data. Some of the values are clearly out of range, but others appear invalid and in-range. :-( So decoding the .spo(x) files without any reference helps - it’s easy to see the O2 and BPM numbers and the remaining are flags and what appears to be a signed integer (but since I only see 0 and -1 i.e., 0 & F not sure). So, I was able to identify a mask for the flags that corresponded to valid data (the vast majority) and therefore that mask might be rejecting some valid data - but not much, so OK. A better definition of the flags for the CMSD50+ would be nice to have. Since I’m importing both the AirSense 11 and O2 meter into Oscar, either the data format/definition of the SD card or the Oscar data files should contain the information I’m trying to analyze. Specifically I need the data values as a time series to cross_correlate i.e. say O2 delayed by tau against various air flow indicators. In the mean time, I’ll be doing statistical analysis of the O2 and cross analysis between O2 and BPM… I’m sure I could get the information from the OSCAR source code, but my C isn’t very good (never was good and long time ago). I will try file data dumping and hope for JSON or somet
