CPAPtalk.com

Thats to bad about your HMO. You are right on about sleeping on your back though. Where your RDI is coming into play, was probably when you were on your back going into your REM cycle. But, you probably already know that I am guessing. you sound much smarter than the average bear when it comes to OSA.

I had a procedure done and when I woke up, the doctor said I stopped breathing twice and snore in between and he intended to fix it. His nurse then handed me I guess what you call a do it yourself sleep study with a recorder sensor to tape on my finger and another under my nose, and told me to bring it back to her the next day. A month or so later, I got a call to "come over and pick up your prescription". That is when I learned about CPAP. I have a couple masks, a machine, and a hose, thankfully no software or humidifier.

john_dozer wrote:Originally I wanted to know if I was leaking bad enough to affect my pressure and verify my pressure.

I got a pressure guage from CPAP.com and videotaped the needle. It revealed the breathing pattern quite well. You could see the needle deflect with each breath. As an apnea approached you could see the needle deflection shallow out to a stop followed by 2-3 large needle motions indicating deep breaths (gasps actually). Perhaps you could put one in the field of view of the video camera.

Relatively decent breathing pattern information and verifies your mean pressures quite well. There's just something about seeing it in analog that doesn't quite translate to digital.

John

Using a manometer fitted with a digital detector (convert the needle movement to digital data) it shouldn't be too
hard to capture that breathing pattern on a PC - it would tell quite a story.

But come to think about it, the pressure & flow transducers in the xpap itself do just that. So in fact if you had your
own device that monitors pressure fluctuations & flows & captures this as fine grained digital data you could start to
emulate the analysis the xpap unit does as it works out what to do & could program the information to be presented
any way you liked.

If you think about it, the algorithms in the xpaps esp the advanced ones, are in effect the best thinking of the manufacturers
research teams. Put another way - these are AI computers (Artificial Intelligence). The best brains of the
researchers & designers captured in code.

For us plebs they hide their algorithms other than the descriptions they put in the patents they file (which is fairly detailed).
If we could do an 'open source' xpap analysis program that captured data from a specially designed (but inexpensive) set
of sensors placed in the air line to the mask (replicate the flow & pressure sensing) we could do our own version of
what they do except we only want the data for analysis rather than using it to change the behaviour of the xpap.

DSM

Using a manometer fitted with a digital detector (convert the needle movement to digital data) it shouldn't be too
hard to capture that breathing pattern on a PC - it would tell quite a story.

I toyed with that idea before getting the analog pressure gauge, but the cost put me off. The analog gauge was about $100, digital recording gauges are quite a bit more. I still might get one if I feel I need to take another "deep dive" into the information.

I hold by breath while I change sleeping postions. My machine scores this as an apnea event. I can also have an apnea event that wakes me up which is followed very shortly by a position change. The APAP report shows the breathing events and the video shows the position change but the clock synchronization just isn't good enough to tell the exact order of events. With the pressure gauge in the video frame you can see it. For an actual breathing event the breathing becomes more shallow or stops, which is followed by a position change. It lets me tell the real events from the bogus ones.

There is also a lot to be said for actually watching your open apnea events in action. I learned that when I have trouble breathing I will shake my head as I try to get air. It's a very distinct motion. Now I can see an apnea event on the tape. For most people sleep is kind of a black box, they go unconscious and then 8 hours later they become conscious again. With the video you can see what happens during that 1/3 of your life.

A recording manometer might be handy. I understand there are some machines out there that let you capture that kind of breath-by-breath detail. I know the machine I have now is seeing all of that, but it doesn't record it.

There is another problem that you have to deal with in data collection, and that's the amount of time it takes to analyze it all. You have to get the information in a format that lets you move through a night's sleep in less than 30 minutes or else you'll never do it.

Although the equipment in the Redneck Sleep Lab is certainly not equal to a professional lab, one advantage that I do have is that I can monitor every night for weeks on end. It lets me see patterns that are not obvious in a one night study and I can observe myself in my normal environment.

TSSleepy, this has got to be right up your alley. Their goal is to help people make homemade EEG devices inexpensively. Here's another group whose goal is to build EEG for under $100. Do let us know if you build one of these things. I'm highly interested in "empirical" science. Hee!

I would love to have an EEG, the lowest price I've seem for one is $1200. Out of my range.

The SleepTracker watch gives you movement information so you can tell when you are sleeping deeply as opposed to sleeping lightly. When I correlate that against the pulse-ox data I can see the sleep cycles. I can tell how long they are and whether they are normal or disturbed.

An EEG would be the "real deal." I could get much more detailed information on what phase of sleep I was actually in. Right now I have to take other indicators and determine the sleep cycles from that.

Be sure to post here is anything comes to market.

LoQ - That EEG site is cool!

My highest "self-test" priority is figuring out when I'm in NREM versus REM. If you look at the ~9-hour pulse oximeter trace below, I have periods of serious oxygen desaturation that look like they are part of my normal 90-120' sleep cycle. I'm assuming they represent my REM stage, but I want to verify that.

The only pre-made EOG I found was way too much ($1055, if I recall correctly). I looked into equipment for an EOG, and one of those should be even simpler than in EEG (less electrodes). I've got my feelers out for a used bio-amplifier. The electrodes and cables aren't too expensive.

Tonight, I'm going to zoom my Handycam onto my pillow. I think I can catch rapid eye movements, if I happen to be facing the camera while those desats happen.

http://www.biopac.com/SecondaryEducation.asp?Pid=4717

Sleep School.

Yay, I was able to detect REM sleep with the camcorder zoomed on my face!

I just "watched the tape" on myself and synched it up with the pulse oximeter results, and my "REM O2 Desaturation Bonanza" hypothesis is confirmed.

I found a period where I was sleeping on my side, facing the camcorder, and was able to clearly watch about 20' of me in one of those desat clusters above.

My eyes would start moving around under the eyelid, my breathing would become labored, and a couple minutes later I would take a couple deep breaths and my eyes would stop moving for a couple minutes. That cycle repeated itself until I finally woke up enough to pop all the way back up to a light sleep and roll over.

Anyway, I just wanted to mention that the "zoom on your face" technique for detecting REM sleep worked for me.

How to see breath by breath with a Purintan Bennett 420E By Becktrev

For my machine, the Goodknight, here is the solution. (Eat your heart out Ozij)
Screenshots!
When the Silverlining is in Data Collection mode, the computer screen display shows a continual graph of the breathing pattern in 5 second segments – a full screen showing circa half-a-minute's breathing pattern.
There is a software called Timesnapper – freeware.
It takes a screenshot of the Data graph every 'x' seconds and stores it as a graphic file. All night if you want. A 100 KB picture every 30 secs – 12 MB an hour. Not much.
Then there is a replay function to check the pictures as a video.
By comparing the time with Silverlining's own Compliance Data, individual Apneas can be looked at and the exact length identified. Look a little harder and you can even see whether it was central or obstructive.

Simple.
The program needs a few adjustments to make it workable. Should anyone have interest, I could explain more.

(No idea why he wanted me to eat my heart out - I jumped for joy when I discovered this thread...)
And here's the promised more.
And more

I downloaded Timesnapper direct from the Owners site - it's safe.
But a tweak is necessary. The screenshots only continue auto as long as there is keyboard/mouse activity. Even my snoring doesn't make the mouse vibrate enough, so when the Data comes from the Goodknight, the screenshots stop being recorded after a while.
The Prog initiates a file called 'settings.ini' in its home directory. Edit this file with the additional line ....
TakeSnapshotAlsoWhenIdle = true
....and it doesn't go into standby

For a night's use, it is necessary to kill screen/power savers of course.

If that's all too complicated, then one probably should not play with the machine anyway.

(Added emphasis mine)
It's excellent.

And - as far as I know - the Sandman Auto has the same data capture capability.
O.