Rescan Statistics v. Detailed Graph Pressures

[welki]

Today, when I viewed last night's numbers, the statistics indicated that my maximum pressure was 16.7. However, there wasn't a point anywhere in the detailed graph that indicated a pressure over (about) 13.

Does anyone know why this discrepancy might occur?

Thanks in advance,
welki

[idamtnboy]

Today, when I viewed last night's numbers, the statistics indicated that my maximum pressure was 16.7. However, there wasn't a point anywhere in the detailed graph that indicated a pressure over (about) 13.

Does anyone know why this discrepancy might occur?

Nope, not me, but I have seen the same sort of discrepancies in some of the other numbers also, like leak rate. I run mine in CPAP mode so I don't worry about the pressure. I wouldn't worry about it. The graph is the most accurate part of the entire data presentation, and as long as you don't see anything frightening there just sleep on it!

[welki]

Thanks, mtnboy, I'm more curious than concerned. I must be getting accustomed to being a hosehead, because a couple of weeks ago, I WOULD have been really frightened!

Even at 16.7 with the AutoSet, the pressure is much lower than the 19 originally prescribed to me if I'd agreed to CPAP. Numbers show that I average under 13, so I'm glad I successfully campaigned for APAP. I figure if I stay below 19, I'm ahead of the game

Now if I could only train myself to exclusively breathe through my nose!

welki

[idamtnboy]

Now if I could only train myself to exclusively breathe through my nose!

Would be nice, wouldn't it? I'm sure you've read enough posts here to realize the ones who do succeed in training themselves to keep their mouths shut during sleep are the minority! That's why tape (my own solution) and other make do answers are so popular!

[AMUW]

Today, when I viewed last night's numbers, the statistics indicated that my maximum pressure was 16.7. However, there wasn't a point anywhere in the detailed graph that indicated a pressure over (about) 13. Does anyone know why this discrepancy might occur?

As mentioned already, one extreme data point isn't important:
- could be what PSG techs like to call an artifact
- note that the machine is able to measure pressure only inside the box; has to extrapolate to what happens in the mask
- I suspect that the values for the curves drawn are not the detailed data measured, rather after some averaging and reduction takes place; except maybe for max / min values

[idamtnboy]

- I suspect that the values for the curves drawn are not the detailed data measured, rather after some averaging and reduction takes place; except maybe for max / min values

All the data displayed in Resscan is taken directly from the data generated by the S9. Resscan does no data manipulation other than scaling. All the summary data, which is what is shown in the statistics page, is generated and written to the card by the S9. The same for the detail data. Resscan is basically nothing more than a presentation program to display the data from the S9.

What does happen, from what I've been able to determine, is the detail and high resolution data files always end on a full one minute increment from the recording start time. This means all detail data generated one second to 59 seconds after the full minute increment is lost. The recording start time for the various data files is not always the same, i.e. the events file may start recording many seconds after the flow data recording starts. I've had instances where the event file shows an apnea several seconds after the flow data graph ends. If the recorded event is real then that means the apnea occurred during the last partial minute of flow that did not get written to the card.

Max/min values are the ones I've seen most often that do not agree with the detail graphs. I also have serious doubts about the accuracy of the 95th percentile figures. In the number of cases where I've looked at the leak data with a spreadsheet, the ss calculation for the median, 95%tile, and max don't agree with the numbers from the S9.

All of the data from the S9 as shown in Resscan should be viewed in terms of looking at an overall general picture, most reliably showing changes in trends. Individual data points, especially the outliers, are not trustworthy. Never make changes in therapy based on one data point.

[AMUW]

All the data displayed in Resscan is taken directly from the data generated by the S9. Resscan does no data manipulation other than scaling. All the summary data, which is what is shown in the statistics page, is generated and written to the card by the S9. The same for the detail data. Resscan is basically nothing more than a presentation program to display the data from the S9.

That sounds logical idamtnboy; if you're ahead of me and actually looked at the detailed data on the BRP file ... and can confirm that the digitized time scale output to ResScan corresponds to the sampling rate in the ResMed spec. But a data reduction step could still take place inside ResScan, if it decides that the plot resolution doesn't need to be as fine as that used in the calculations. Which is just a suspicion; I don't know that for fact.

Welki, after my last post, out of curiosity I went back to my Pressure ResScan plots for recent days, and compared to what appears in the Summary data for the same day: the max value seems to agree everywhere.
One possible reason for your disagreement: are you sure your Summary data corresponds to the same one night time span: same start /end time, same duration? I happen to remember similar complaints from people who don't have their ResMed internal clock set properly, or who sleep past noon ... or who may live in the Polynesian Islands who just decided last week to move their clock into another time zone

[welki]

Yes, I'm sure that the date and duration are the same, the clock is set correctly, I woke up at 6:00 a.m. and I don't live in Polynesia. Oh, and yes, I turned the iron off, locked the back door and made sure the dogs peed before I loaded them into the truck

Not worried about the anomaly, since this machine is on a "trial run" until Tuesday, when I purchase my own machine. (But I will mention this to my RT.)

Thanks to both of you for your replies.

welki

[terio2]

Because of a sore throat I used the S9 apap for only 4hr and 40 min. I downloaded and only have the main pages no details. I want to know what events I had like usual. Anyone know why rare occasion it doesn't give detail. This happened before with the machine on longer.