[Previous Months][Date Index][Thread Index][Join - Register][Login]
[Message Prev][Message Next][Thread Prev][Thread Next]
Re: [IPk] Blood Charts and techno-utopia
>This type of approach can not be too far off. There is a chap at
>Robert Gordon's University in Aberdeen who is
>modelling type one diabetes - his work is still at the very
>embryonic stage though. At this stage it is passive - he is gathering data
>on bg's, insulin usage, CHO intake, exercise and modelling it all. I
>participated in the first phase of the research but then started on my pump
>so could not go any further His young daughter is diabetic and I feel that
>he his hoping that there will be a magic way to forecast the whole thing - I
>hope he is right!
A chap called Eldon Lehmann has done something like this too. See
http://www.2aida.org and more specifically
http://www.2aida.org/aida/technical.htm , which is a fairly detailed look
at the history and current situation with regard to simulating diabetes.
But we all know how enormously complex it is to predict what our own
diabetes is going to do. Even worse, it varies so much from person to
person. One person has a large dawn effect, another person has none. Why?
My concept does not necessarily need to model the internal functioning of
each person's body. Weather forecasting is done by looking at the current
and recent readings, looking for similar situations in the past, and what
happened last time is the best prediction of what is going to happen next
time. You don't need to know why an occluded front here predicts rain there
- just that it does. Similarly, a computer may have spotted that when
you've eaten a large dinner with red wine, your carb/insulin ratio the
following morning drops a bit. It can use that info to optimise your
breakfast bolus, in a way that I don't think I would have the time or
inclination to do.
The longer you use the system, the better the computer gets at spotting the
little quirks of your body, and the less information it needs to give
helpful advice. The big question is whether there is a window in which the
small amount of info you can be bothered to give is enough to generate
meaningful advice. That we may only be able to find out by trial.
mailto:email @ redacted
for HELP or to subscribe/unsubscribe, contact: HELP@insulin-pumpers.org
help SUPPORT Insulin Pumpers http://www.insulin-pumpers.org/donate.shtml