Simon James wrote: >> I also bought an OCR program recently: >> IRIS Readiris Pro 11 ( http://www.irislink.com/ ) >> >> but it hasnt arrived from Belgium, dont hold out much hope, have written >> twice etc >> >> I dont think that OCR is a good long term solution because it could imply >> tampering down the track, but it would enable us to Copy&Paste into a >> template >> for smaller storage for each scan/import. >> > > Hi Fliss, > > I think scanning to PDF (with a hidden OCR layer behind the image) is the > only way to go (I think IRIS does this). > > Trapping just an OCR interpretation or scanning to TIFF/JPG (or a PDF sans > OCR layer) are fairly big compromises IMHO. > > I'm actually about to start my feature article for the Feb edition of Pulse > IT under the wanky working title of the "war on paper". I intend to cover > options for retrospective and ongoing scanning, and hope to compare how a > few popular EHRs deal with these documents once created. > > Scanning will constitute only 1 sub topic, and I hope to encompass: > > - Online claiming > - Electronic faxing > - Secure messaging > - Offsite options (PDAs, tablets, remote access) > - Filmless imaging > > I'd love to hear from people who are getting the most out of these > technologies, as well as people who have had difficulty realising the > potential. > > All the best, > Simon > I have a problem with pdf bloat. Not so much the bloat of the file size as the bloat of Adobe Acrobat, which seems to have delusions that it is an operating system. Also, there is no good open source pdf reader on Windows.
Most of the data I get is either atomised pathology data, which should never be scanned for OCR in my opinion, or text. For text my needs are simple and are limited to fonts, bold, italics and underline. I don't like tables in text documents. I detest pictures. For me HTML would be acceptable but I recognise a more general format is desirable. This should be Open Document Format since this is an open standard. The docs for ODF are 700 pages long. You could use the MS XML format standard but the docs for that are 4000 pages. GPs work at a higher conceptual level than specialists. They love telling us the details of the pelvic cavity exploration, or the anatomy of the meniscii, the texture of the colon or the pulmonary arterial pressures. While these are fun, I really need the pathological and clinical implications of their findings. For this simply formatted text is best. As far as security is concerned WYSWIG does not apply. I can fairly easily manipulate any document format to look and feel exactly like the original. Tim has a little man in the back streets of Manilla who can do they same for you if you want a paper copy. The only thing that has any scientific validity is a cryptographic hash. There were recent discussions on the list about how to send that in the HL7 message and general disappointment that no EHR in common use in Australia saves that data from the HL7 message. Personally I think the whole thing of altering medical documentation is a furphy. The only thing likely to save you in court is the extent of your genuine documentation. If you are foolish enough to alter one iota, you are gone. If I were a medical litigant the first thing I would do is subpoena the current and backup records of the patient's file taken one day and one week after the episode in question. I would also subpoena the same documentation from the originating medical party. If one used Horst's gnotary service there would be no question of the records being a true and accurate representation of the medical record at the time. However, the gnotary service is not, and will not, be widely used until such time as we are sued more regularly and expensively. These issues have been discussed on the list intermittently over the last few years. Searching the archives will turn up a more in depth discussion. HTH. David P.S. Is Pulse a Joomla site? _______________________________________________ Gpcg_talk mailing list [email protected] http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk
