[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnumed-devel] document control

From: J Busser
Subject: Re: [Gnumed-devel] document control
Date: Sun, 10 Oct 2004 17:58:58 -0700

At 7:19 PM +0200 10/10/04, Karsten Hilbert wrote:
IOW, if the path results *appear* to be a document - store them
in the document tables. If the appear to be values put them in
the measurements tables.

Eg. a document present *information* while values present

I think we are here talking about documents (or "files" as when a series of paper originals have been scanned into single or multi-page TIFFs). Such "documents", in their original form contain "information" as Karsten puts it, and this information may or may not be useful to convert using optical character recognition (OCR) and/or parse into its component data items.

On the one extreme, there are labs whose support of an EMR is so primitive that they really only send an ASCII file which, if it were properly formatted, may be able to be printed to look similar to their traditional paper report. The format can be so idiosyncratic that for anyone to create a parser/translator would cause the person to pull their hair out and one can never know when the source will choose to modify their format as they use no "standard". In such circumstances, where recipients choose not to bother with a parser/translator, the only practical choice is to retain/archive it as a document.

On the other extreme there are labs that do not "send" or "push" "documents", they only make their results database available to be queried ("pulled") by the fetcher/importer software. It is true that the data stream could be saved as a "document" but as it is likely human unreadable except with difficulty it would not seem worth considering it a document.

Much thought could be required on narrative documents, which could contain chunks of data of interest (problems lists, family histories, medication lists, other). However, until such documents adhere to an interchange standard, they are probably only worth retaining/archiving as "documents". Users may wish to copy/paste portions into appropriate GNUMed records.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]