Mailing List Archive Mailing List Archive
Nick!
You mail feed is confused. I am already established as one who avoids
decompression diving and I said so in my introduction.
>> >What about a next generation diveputer with a doppler-based bubble detector
>> >connected with a neural nettwork which estimate the future probability of
>> >DCI? I'm serious neural nettwork can estimate chaotic timeseries and the
>> >probability of bubble formation is (I gess) an ODE.
Not me (you can tell by the little '>'). Since this posting someone who
knows more about neural networks explained why neural networks in particular
wouldn't be very usefull for modelling DCI. My view is that some techniques
like fuzzy logic and chaos theory might, but I don't know enough about them.
Know any more about these methods?
>> I prefer to stay within the envelope for no-decompression diving myself.
>> I've got more to live for.
>On one of the other lists, I'd let this pass. On this list, this is
>an incredibly stupid statement, which makes no sense in response to
Personal abuse?
>the above brainstorm about an improvement in dive computers. Why do
I'm all for new ways of looking at the data, I still think your newsfeed is
confused...
>you think that the above implied diving beyond the No-stop limits?
Allow me to clarify, I like to surface straight to the boat. I stay within
the no-deco limits. If you want to hang in the water decompressing, go
ahead. Around here you are very likely to be eaten by something.
>And, of course, every dive is a decompression dive.
I think I got flamed once or twice before for asserting just that (I still
assert it). Am I now being flamed becasue it is perceived that I don't?
>I think that this computer might be most useful as an after dive
>estimator. I'm not sure exactly when bubbles form, but I expect that
>the most significant results might be after returning to sea level.
Re-read the post on why neural network techniques might not be appropriate here.
>However, I think that the real problem might be that I, for one,
>wouldn't want to be one of the participants in the neural network
>training. After all, one assumes that there are a large number of
How do you decide in an empirical way who has DCI? Note also that I'm using
a different terminology to you. You are using the old 'mechanistic
terminology' which isn't very useful once you actually have symptoms which
need treating. I guess these are the ones people want to know about. Who
cares about symptoms that don't need treating?
>cases where there would be no DCS, and another large number of cases
>where DCS would be quite probable. There would also be a large number
>where DCS probability would be marginal, and the training would
>require people to be bent so that the network would learn how to
>discriminate.
A fantastic concept. What criteria for discrimination would you use? As far
as I am aware there are no objective criteria for deciding whether or not
someone has DCI. Even the presence of intravascular bubbles is often
without symptoms.
>Finally, I'm not sure of the value of the statement, "Your bubble
>profile indicates that you have a 34.7% chance of developing DCS."
>What do I do? Head for a chamber one time out of three? Carry out an
>omitted decompression procedure if I am not showing symptoms? Breathe
>100% O2?
You should ask whoever said this. I don't understand it either.
Have a nice day.
/Rat
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
shelps@ac*.ma*.ad*.ed*.au* | Stephen Helps Ack! ___/|
FAX (08)232-3283 | Anaesthesia & Intensive Care \O.o|
Voice (08)224-5495 | University of Adelaide =(___)=
| ADELAIDE, 5005, South Australia U
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In ancient times the dinosaurs
did oft' another vex,
And charging at each other
made tyrannosaurus wrecks.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Navigate by Author:
[Previous]
[Next]
[Author Search Index]
Navigate by Subject:
[Previous]
[Next]
[Subject Search Index]
[Send Reply] [Send Message with New Topic]
[Search Selection] [Mailing List Home] [Home]