On Wed, 6 Jun 2001 00:12:32 Alexander Chelnokov wrote:
>Hello Tim,
>
>Tuesday, June 05, 2001, 9:31:50 PM, you wrote:
>
>TC> with the full text searching capability across the entire set of
>TC> records I believe a practice can build an enormously useful repository
>
>Text search is great of course, but it is only one option from
>required for medical data analysis. And, or, not, more than, less
>than, between, inlist, isempty and so on are also basic. Search on
>on-line constructed expressions like admission_date-discharge_date>5
>would also be necessary.
These are already provided by SQL-capable database management systems.
That is why OIO comes with these capabilities with very little additional code.
Zope can do some of this but require more coding. Tim knows more about Zope vs. SQL so
I look forward to his remarks.
>TC> Can you duplicate 'near' searches in context, in SQL? With OIO you
>
>There are search tools in bibliographic systems which allow something
>similar with defined number of other words between search terms. But
>our tasks rarely are limited with search only. Found stuff has to be
>analyzed. Something like across the hospital calculation of safe days
>in blondes less or equal 25, with waist less than 65 cm and who lives
>not farther than 25 km from my office. Or for instance find answers to
>questions like "Do smokers have longer healing period than non-smokers
>in open grade III tibial fractures treated with external fixation?"
>"How quality of life changes after different managing programs and are
>differencies between and inside groups significant?"
Yes! This is exactly what the OIO system is designed to do.
>TC> What information do you see that is missing from FreePM that would caue
>TC> you to make that statement?
>
>Except discussed problems of i18n <:->, IMHO it would be great to
>have ad hoc query tools (in FreePM as well in OIO).
OIO already has online analysis capabilities. And/Or/Not in browser-accessible
drill-down screens. Date operations still require messy SQL writing (through editing
Zope SQL methods). This should change over time. :-)
> With more or less
>standard design - user can define search conditions (see above ;),
>output fields,
Done.
> sort order,
Not yet. (Sort by field name only for now. How else?)
> output format (comma delimited text or
>spreadsheet, PDF/graphs, etc)
We have comma delimited text and XML for now.
>and, optionally, basic calculations
>(average with standard error, min, max, count etc).
Done.
> Actually i would
>be happy having any table output which i could open with Excel and use
>its Pivot Table, base statistics and chart capabilities.
Done - you can import comma delimited text directly in Excel. I do this all the time
with OIO-collected data.
>Or process
>the table with a statistical package like SPSS or Statistica.
Done - you can import data into SPSS.
Best regards,
Andrew
---
Andrew P. Ho, M.D.
OIO: Open Infrastructure for Outcomes
TxOutcome.Org (hosting OIO Library #1)
Assistant Clinical Professor
Department of Psychiatry, Harbor-UCLA Medical Center
University of California, Los Angeles
Join 18 million Eudora users by signing up for a free Eudora Web-Mail account at
http://www.eudoramail.com
_______________________________________________
FreePM Support List
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/freepm-discuss
