Ill take a stab at this. Answer is for a lot of things they cant.
They do a decent job of telling CA from OAs, flow limitations. RERAs they make a guess at.
If the data is based on flow and backpressure or lack of it, they do a good job.
Without wrist pulse ox monitors etc they cant tell your puse rate O2 Sats etc like the lab can nor read your brain waves.
All in all unless you have something unusual going on that needs brsin wave data etc, they give the average pap user all the data he or she needs to optimise therapy or tell if something is obviously wrong.
The machine have air flow sensors
When air flow drop 80% below the baseline for 10 seconds or more, an apnea is scored
Of course the machine does not have EEG, so an apnea can be scored while you're awake which are not scored in the sleep lab PSG
There is more to it ...
A person breathes in about 20 times a minute. A good sample of that sort of timing - to get a fair view of air flowing in and out or a person - requires a few hundred samples per breath.
Since memory is cheap these days, machines typically sample more than they need since it alleviates several sampling theory byproducts like aliasing. In my machine, it records about 150k bytes per ten minute period, or 15k bytes per minute, or 250 bytes per second. A byte is 8 bits, so there are 2,000 bits of data being saved. Unsure if compression is being used, but probably is, so this is a lower estimate.
Today's machines with airflow and pressure sensors have analog to digital converters that take those values and convert them for the computer (embedded microprocessor) to process. The MCP runs at several MHz, so dealing with 250 byte a second real-time is easy enough. Though, I suspect the actual sample rate of the sensors is much higher than 250Hz, more likely around 1kHz to 10kHz and the extra data used to improve computational and analysis accuracy. Much of this work is done in specialized hardware designed for the purpose... the math and programming is called DSP (digital signal processing) and is usually bundled on a special DSP chip or FPGA. The MCP most likely simply routes the data between sensors, DSP and memory. The DSP chip probably has the proprietary algorithm to control the air pump to provide the correct pressures at the correct times.
These days the FPGA's being sold have sections that can even be programmed with MCP cores, ADC's, MUX's, and more; it can all be bundled into a single chip. All that's needed outside the core chip is power devices to drive the motor.
To get a flow rate measurement, one only needs two pressure sensors in a tube, and some tinkering with diameters and positions.
I've kindof gone off the deep end in detail... but think iPhone for size and capability; and consider that an APAP is doing is much much simpler work.
Lots of good info above. Another wording.
Change fan speed to control pressure.
Math and magic.
Graphs and data.
Much the same as what happens in a sleep lab, but with fewer sensors and a lot more computer processing of the data to determine AHI, CA, etc.
The CPAP machine suffers from not having an EEG sensor and chest belt. It can't tell whether the patient is awake, and can't be as accurate about central apneas.
The data capable CPAP still provides a LOT of very useful data. It's also data from your real therapy, every night. It's a bit like having a diabetic check his own blood sugar daily. it's not as good as a full set of lab tests, but you'd be a fool not to use it, since it's freely available and very useful.
Get the free SleepyHead software here
for information on the main alternative to CPAP.
If it's midnight and a DME tells you it's dark outside, go and check it yourself.
09-13-2015, 10:45 AM
(This post was last modified: 09-13-2015, 10:48 AM by justMongo.)
No one is going to get into the Nyquist–Shannon sampling theorem and Nyquist frequency? [/sarc]
INFORMATION ON APNEA BOARD FORUMS OR ON APNEABOARD.COM SHOULD NOT BE CONSIDERED AS MEDICAL ADVICE. ALWAYS SEEK THE ADVICE OF A PHYSICIAN BEFORE SEEKING TREATMENT FOR MEDICAL CONDITIONS, INCLUDING SLEEP APNEA. INFORMATION POSTED ON THE APNEA BOARD WEB SITE AND FORUMS ARE PERSONAL OPINION ONLY AND NOT NECESSARILY A STATEMENT OF FACT.
Wow! A lot of the info from the replies is far too complicated for my foggy brain to comprehend. I get the general gist of it though and it makes sense. Good to know and thanks to all for the detailed and helpful replies!!! This forum has so many helpful and knowledgeable people!!!
As to your question " able to tell my sleep Doc why I need a fully data capable machine for my loaner period instead of the brick "
There are many settings that are referred to as "comfort" settings that help a patient tolerate the main therapy. If a patient will tolerate the therapy, they will use the therapy.
A data capable machine allows the patient to see if the changes they made to these comfort settings actually improved their therapy tolerance. Most things regarding CPAP therapy happen while we are asleep, so having a machine that can tell us what happened while we were sleeping can in some cases be the only feedback we get and help us to adjust the comfort settings so we tolerate the therapy better.
Many patients stop using CPAP, even though the therapy is reducing their apneas, because they just cannot get comfortable using the machine. A data capable machine gives the patient the best chance of being successful.