I'm backtracking to an old item so as to respond to this email from Syan
as it implied there was some role we could serve.
I don't find syan's proposal for this research so readily achievable.
Firstly we would have to have access to a collaborators data and DB
environment to do anything at all. That is not an easy thing to arrange.
Secondly we would then have to come to terms with the DB software and
the schema for the application - a lengthy task. Thirdly the matter of
what analytics you provide is a very open question. Let me describe what
we are doing on this quarter.
Currently we are working at three sites learning how to use 3 different
DBMS and naturally 3 very different schemata. The sites are:
Westmead Hospital & SWAHS on the CARDS system - a specialist Cardiology IS
SEALS (SouthEastern Area Laboratory Services) - part of SES&I AHS
working on their ONTOLAB system
RPAH -ICU Carevue system.
In each case we are doing what we call "Forensic Data Modelling", that
is, reconstructing the data model and figuring out how to program in the
local language to get at the data. We are on our 6th project at the at
the RPA, 2nd Project at Westmead and first at SEALS. It took 5 projects
to crack the problem at RPA, we've been little faster at the other
sites. A project is 1 student semester of work.
At the Westmead and SEALS we are just trying to rescue the data from
deteriorating ISs and to help map a pathway to new systems. We have been
asked by NSWHealth to provide a report on this work so that they can
consider how to tackle the problem across the whole state.
At the RPA we are tackling a wider problem which we call Generalised
Data Analytics, that is how do you answer any question for which the
answer is in the database - this is more than just getting the data into
an ehr and learning how to use some statistical methods. This work is
part of a wider project to build an Intensive Care Real-time Audit
System (ICRAS) for evaluating patient care continuously.
Yes we are happy to collaborate with members of the list but doing the
work to produce results is not something that can rattled off so easily.
cheers
jon
syan tan wrote:
unfortunately, that's a big if. What's the rate of uptake for
computerized records nowadays : are almost all of the potential emr
converts already converted/locked-in with the current crop of cottage
industry applications ? Maybe someone should try to use some statistical
method (like monte-carlo or bayesian filters) to look at the data
contained generically ( just need to know how to access
4D databases, microsoft SQL, foxpro, and possibly firebird/interbase )
look for common patterns of data layout/ character/word sequences , and
attempt extractions into a common ehr schema format. Doable ? say for
a/some university researcher(s) ?
On Mon, 2007-04-23 at 11:22 +0800, Richard Hosking wrote:
Yes after a few months on the Rails thing I have given up for the moment
Too many other things
But as you say later an OSS platform that just started with the basics -
billing/front desk/basic records EHR would certainly get things going if
there was a reasonable number of users and data. You could build all
sorts of things on top of it.
R
David Guest wrote:
David More wrote:
Hi David,
I think there are some groups doing pretty smart decision support and
doing it pretty well. Certainly the literature on CPOE shows the
better implementations of that are making a difference.
Medical Object has some really fantastic standardized stuff in the
Cancer Domain also.
That's good to hear. I wasn't aware how far MO had come since I last
suggested to Andrew that he should be a player in the EHR domain
(since he understood the comms area).
However, like Argus eprescribing and JJ's ICP initiative, the limiting
factor is the closed source of the EHR. It basically shuts off DS
experimentation from those who are willing to dabble.
Don't be quite so despondent - there are islands of hope out there!
Bored more than despondent, David. There are lots of fascinating
things happening in the world and the gross inefficiencies of the
medical IT industry are a mere triviality.
Do you play golf?
David
_______________________________________________
Gpcg_talk mailing list
[email protected]
http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk
_______________________________________________
Gpcg_talk mailing list
[email protected]
http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk
_______________________________________________
Gpcg_talk mailing list
[email protected]
http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk
--
Jon Patrick
Chair of Language Technologies +61-2-9351 3524
School of Information Technology
Faculty of Engineering & IT
University of Sydney
begin:vcard
fn:Jon Patrick
n:Patrick;Jon
org:Faculty of Engineering;School of Information Technologies
adr;dom:;;University of Sydney
title:Chair of Language Technology
x-mozilla-html:FALSE
url:http://www.it.usyd.edu.au/~jonpat
version:2.1
end:vcard
_______________________________________________
Gpcg_talk mailing list
[email protected]
http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk
