Entropy in this sense is from information theory.
I am not certain of the full history but I came across it in Claude
Shannon's work which is available in pdf format on the Lucent technology
site.
While his work was primarily concerned with elimination of noise in
communication, it has, I believe, potential application in understanding and
quantifying the medical decision process.
There are a number of potential extensions - for example one might use it to
define data redundancy and of clinical utility.
You might like to try http://www.math.psu.edu/gunesch/Entropy/infcode.html
which includes a link to Shannon's paper but that link is in post script
format. If that is not suitable or you cannot find it on the Lucent site I
could append a copy in pdf format to an e-mail should you wish.
I am not aware of it being used in quantification of medical decision
making.
Trevor Tickner
Norwich
> -----Original Message-----
> From: Robin Marks [SMTP:[log in to unmask]]
> Sent: 22 November 2001 15:39
> To: 'TICKNER TREVOR (RM1) Norfolk and Norwich NHS Trust';
> [log in to unmask]
> Subject: RE: Decisions, decisions
>
> > -----Original Message-----
> > From: TICKNER TREVOR (RM1) Norfolk and Norwich NHS Trust
> > [ <mailto:[log in to unmask]>]
> > Sent: 21 November 2001 17:31
> > To: [log in to unmask]
> > Subject: Re: Decisions, decisions
> >
> ...(snipped)
> > If, however, one divides the population on the 100 barrier
> > there will be two
> > groups. That below 100 will have an entropy of 0.045 while
> > that above will
> > have an entropy of 0.918. Uncertainty is clearly much less in
> > the 'excluded'
> > group. Mean entropy is 0.7 but most of the information gain
> > has been through
> > exclusion.
> >
> ...(snipped)
> --------------------------------------------------------
>
> Trevor,
> Could you explain how you calculate 'entropy' here ?
> Is this the same as entropy in chemistry and physics
> and information theory ?
> Has it been used in the context of clinical decision
> analysis before ?
>
> Robin Marks
> Consultant Chemical Pathologist
> Calderdale Royal Hospital
>
>
>
---------------------------------------------------------------------------
The information contained in this e-mail is confidential and is intended
only for the named recipient(s). If you are not the intended recipient you
must not copy, distribute, or take any action or reliance on it. If you have
received this e-mail in error, please notify the sender. Any unauthorised
disclosure of the information contained in this e-mail is strictly
prohibited.
|