Is learning device just a dream?

Out of curiosity, has anybody ever seen a learning APAP?

I think it shouldn't be that difficult to add some learning in the algorithm such that
the machine actually used the recorded data for anticipating the behaviour of the curves and
trying to optimise the quality of sleep. Kind of more global optimisation instead of just reacting
to current events to "personalise" the response. The learning info could then be saved, reloaded
and cleared.

Respironics tried that with their latest M series machines, and well they seem to have literally nailed it...

...because just like the patients it attempts to treat, it seems to have learned to forget things just like the patient,
and forget what it did last night

In other words: Dream on (if you can)?

If you mean leaning from night to night - I don't think so, and I think machines had better start nights anew.

However, all machine use your breathing patterns and average the results for the past few breaths before they identify your present breath as normal or flattened, or obsturcted, etc.

O.

Typos edited....

So a learning device is just a severely apnetic and desaturated dream .

Since a person typically sleeps differently each night, I don't think it would be a good idea to try to replicate the previous night's sleep patterns......especially if they were NOT good.


Den