SleepyHead 0.9 beta sneak preview..

[jedimark]

I'm keeping busy getting the SleepyHead 0.9 beta ready, but there's a couple of things I need some input on..

I thought I'd post a couple of screen-shots to see if anyone less foggy has any suggestions as to what data is missing from these pages..

They are of the new Statistics page, and the new sidebar. The sidebar can now be accessed from any page, and toggled on and off with a single keypress to reclaim screen space.

The statistics are shown differently depending on machine type. If it's all CPAP data, just the pressure is shown, if it's BIPAP, it shows Min EPAP, Max IPAP, and the 95% values for the two..

Records area of the sidebar is for showing good or bad achievements, and provides a link to hop the data in question in daily/overview tabs..

The new Bookmarks tab provides a quick link to the days that have areas Bookmarked, ie, click on a date, and it opens in the Daily tab.

Can anyone think of any text info/statistics worth adding to either this statistics page or the Records area in the new sidebar?



The coloured highlights in the "Changes to Prescription Settings" show the best (in green) and worst (in red) setting so far, based on AHI.

Underneath this one shows a list of Machine information.

[robysue]

That statistics page looks really nifty JediMark.

Things I can think of to add are:

• same kind of statistics for RERA and Flow Limitation indices that you have for the AHI
• median and 95% snore numbers
• total leak numbers for those of us using machines that report total leak instead of unintentional leak AND/OR a short explanation of how the leak rate numbers are computed in SH for machines other than the Resmed S9. In other words, including a note about Leak rate computation that is similar to the note you made about the summary AHI numbers being computed differently here versus in the overview chart would make these numbers more meaningful.

And thanks for adding those 95% numbers on IPAP and EPAP for those of us using BiPAP Autos.

I'm really looking forward to the 0.9-beta version!

[Perrybucsdad]

I love it Mark.... I'm sure I could think of a hundred different things to add to the stats page, but I think that is a great start. It may be nice to allow the user to customize what they have on there and the timeframes reported, but that could always be a later version.

When will v0.9 hit SF to download?

[ozij]

Looks great, Mark.

I'd add number of sessions to the statistics, and an option to discount sesssion in cases when you know you were awake, and the bunch of apneas there is skewing your data. When I used Puritan Bennett's Silverlining with its great session by session export I simply dropped any seession shorter that 60 minutes from my calculations.

[Breathe Jimbo]

If you are going to add many more types of information, please make the types optionable, so that we can keep the Statistics page as a very simple highlights page with only the key data that we want to see there. All the other stuff can already be seen in more detail on the other pages. It is nice to have this one simple page with only key statistics, viewable at a glance. No sense in making it a pale imitation of the other, more detailed pages. IMHO, the focus here should stay narrow and focused on efficacy only.

[jedimark]

I love it Mark.... I'm sure I could think of a hundred different things to add to the stats page, but I think that is a great start. It may be nice to allow the user to customize what they have on there and the timeframes reported, but that could always be a later version.

When will v0.9 hit SF to download?

Customizable like that is Easy to say... not so easy to accomplish, especially in a graphical sense. SleepLib backend has loads of channels and calculations that are relatively easy to access.

Perhaps an editable XML file to allow for user customization? Obviously this limits it to advanced users, but then you can always share your scripts..

Anyway, should be out in a new few days. maybe not with that ability just yet..

[Lizistired]

Is there a way to add "EPR/FLEX setting" to the changes to prescription settings?

[jedimark]

If you are going to add many more types of information, please make the types optionable, so that we can keep the Statistics page as a very simple highlights page with only the key data that we want to see there. All the other stuff can already be seen in more detail on the other pages. It is nice to have this one simple page with only key statistics, viewable at a glance. No sense in making it a pale imitation of the other, more detailed pages. IMHO, the focus here should stay narrow and focused on efficacy only.

I rather agree here.. I don't want to be too busy by default... I think this is another case for the xml file thing..

Maybe could have several "profiles" to choose from.. (Beginner, Medium, Advanced, etc...) in a drop down in preferences, which basically select from a default set of XML templates.

[archangle]

Average and median leaks is going to confuse a lot of people.

Should 90% leak be on the report?

[jedimark]

Is there a way to add "EPR/FLEX setting" to the changes to prescription settings?

My machine has three modes (+off) of pressure relief, each mode having 4 levels. I'm not sure if EPR is similar.

Adding the modes is definitely a good idea, as the user "can't" change this without alter prescription settings.

As for the levels I'm not sure, as it's usually user adjustable. It will make the list rather long, and will dilute the results quite a bit.

There doesn't seem much point to me comparing single days worth of data, which is why I added the threshold here to ignore them in the best/worst RX calculations.

[jedimark]

Average and median leaks is going to confuse a lot of people.

Should 90% leak be on the report?

Average and median are quite different statistics.. but they don't both have to be there..

I was basically showing the new Percentile engine can calculate any percentile for any data source now..

What I'm really looking for is a list of the most important things to show by default. Customization can come later..

[jedimark]

Looks great, Mark.

I'd add number of sessions to the statistics, and an option to discount sesssion in cases when you know you were awake, and the bunch of apneas there is skewing your data. When I used Puritan Bennett's Silverlining with its great session by session export I simply dropped any seession shorter that 60 minutes from my calculations.

Down the bottom where the sessions are listed in Details view, I'm in the middle of adding a switch to turn sessions on or off.

The ignore slider in preferences already stops them from being loaded.. (Try it and see.. I personally see no point in keeping <1hr sessions, as I doubt anyone really gets much valuable sleep during that time)

When the on/off buttons are completed, the sessions ignored by this slider, will simply be set to off, and they will still display in this list.

[idamtnboy]

Even though I don't use SleepyHead, at least not yet, you're doing a super job. The statistics page looks good. My only thought is I would suggest you round off all, or most anyway, numbers to 1 decimal place. To me, taking them to three decimal places tends to muddy up the larger view of the numbers, and gives a sense of exactitude that I don't think reflects reality.

[Perrybucsdad]

I could have sworn I saw that I saw 0.9 on SourceForge last night, or was I dreaming?

[jedimark]

I could have sworn I saw that I saw 0.9 on SourceForge last night, or was I dreaming?

Mac users sorta have a rough early version.. It's missing heaps of stuff though..

It's an unofficial release. My best advice for mac users is to wait.. (or build from source )