shan@mcf.UUCP (Sharan Kalwani) (07/13/88)
Fellow Netters,
A group here has been asked to look into the possibility of taking
boxes and boxes of medical records and documents and converting them
into some form of archival storage. They initially considered
microfilming them but I feel that using the document scanners with
perhaps WORM devices would be a better alternative. However, we do not
have any expeirience in this sort of actitivy at all.
What experiences do folks have using the scanners out there?
What kind of software is necessary? What is the minimum hardware
necessary say for a souped up PC or 386 system in terms of graphics
and image hardware (of course coupled with well regarded text software
which can incorporate these into word porcessing programs)? What companies
have good products and how much do they cost?
We are willing to delve into this seriously. Some guidance will be
most appreciated.
--
usenet ..!{uunet!umix, pur-ee!iuvax, ucbvax!mtxinu, rutgers!psuvax1}!mcf!shan
internet: shan%mcf.uucp@umix.cc.umich.edu shan@mcf.uucp
"The answer is UNIX: now what was the question?" -- title of a technical
report published by some one in some university somewhere in the UK.msmith@topaz.rutgers.edu (Mark Robert Smith) (07/14/88)
The office I work in is looking for a system based on an IBM PC or
clone to mechanically read text ridden sheets and collect data as a
tab house would normally collect the data. Can anyone give me
pointers to scanners or software to do this? We'd additionally like
to interface this data with SAS on a Novell network.
thanks
Mark
--
Mark Smith (alias Smitty) "Be careful when looking into the distance,
61 Tenafly Road that you do not miss what is right under your nose."
Tenafly, NJ 07670 {backbone}!rutgers!topaz.rutgers.edu!msmith
msmith@topaz.rutgers.edu Bill and Opus in '88!!!wew@naucse.UUCP (Bill Wilson) (07/14/88)
From article <Jul.13.18.20.55.1988.5348@topaz.rutgers.edu>, by msmith@topaz.rutgers.edu (Mark Robert Smith): > The office I work in is looking for a system based on an IBM PC or > clone to mechanically read text ridden sheets and collect data as a > tab house would normally collect the data. Can anyone give me > pointers to scanners or software to do this? We'd additionally like > to interface this data with SAS on a Novell network. > Try calling Flagstaff Engineering. 602-774-5188. Try talking to Gary McClellan. He is the head of the scanner software area and should be able to help you. Do not talk to Connie. I've had problems with her. The scanner software that they produce is reportted to one of the better packages. -- Bill Wilson (Bitnet: ucc2wew@nauvm) Northern AZ Univ Flagstaff, AZ 86011 {These views are mine and do not necessarily reflect those of my employer}
evas@eurtrx.UUCP (Eelco van Asperen) (07/15/88)
in article <424@mcf.UUCP>, shan@mcf.UUCP (Sharan Kalwani) says: > > A group here has been asked to look into the possibility of taking > boxes and boxes of medical records and documents and converting them > into some form of archival storage. They initially considered > microfilming them but I feel that using the document scanners with > perhaps WORM devices would be a better alternative. However, we do not > have any expeirience in this sort of actitivy at all. > Before you dive in and store it all on magnetic or optical storage, you might want to look at the durability of the media concerned. I recently saw a documentary about a big library in the UK facing similar problems and they decided _not_ to use any kind of magnetic (not sure about optical) storage just for this reason. I'm no expert in this field; just thought you might like to know. Eelco van Asperen ------------------------------ Erasmus Universiteit Rotterdam T H E N E T H E R L A N D S ------------------------------ uucp: evas@eurtrx ..!mcvax!eurtrx!evas earn/bitnet: asperen@hroeur5 ------------------------------
davidsen@steinmetz.ge.com (William E. Davidsen Jr) (07/15/88)
In article <424@mcf.UUCP> shan@mcf.UUCP (Sharan Kalwani) writes: | A group here has been asked to look into the possibility of taking | boxes and boxes of medical records and documents and converting them | into some form of archival storage. They initially considered | microfilming them but I feel that using the document scanners with | perhaps WORM devices would be a better alternative. However, we do not | have any expeirience in this sort of actitivy at all. We evaluated scanners and chose the HP Scanjet, as did InfoWorld last week. It will run on an AT class machine quite well, and has a reasonable set of features. Output to a LaserJet is really good, and quite fast if you use a *parallel* interface. This is reasonably labor intensive, and you may want to at least look at the system Plexus makes to do just what you want. It was demoed at the UNIFORUM in 87. One of my friends used to sell them and he won a contract based on cost vs microfilm, so systems like this are cost effective if you have a big application. Warning... the state of character recognition on PCs is such that you will probably not be able to convert to ASCII, since the error rates which are normally encountered are not acceptable for medical records (not mine anyway). I would be pretty happy with the Scanjet stuff, though, I done a lot of stuff and been very happy with the results. -- bill davidsen (wedu@ge-crd.arpa) {uunet | philabs | seismo}!steinmetz!crdos1!davidsen "Stupidity, like virtue, is its own reward" -me
WMartin@SIMTEL20.ARPA (William G. Martin) (07/18/88)
Since the subject of scanners came up, I'd like to see anyone on the list who has had experience with hand-held scanning units, or has done lots of input from random published sources, like books, newspapers, and magazines, post their comments and opinions. The main need I have for a scanning device is for one that will read in arbitrary print or font types from things that you cannot feed through a sheet-feeder mechanism designed to process 8 1/2 X 11 paper, and most scanners I've seen advertised or demonstrated are the latter. I've got ads from a company called SORICON (Boulder, CO -- 1-800-541-SCAN) which makes a little mouse-like unit you can run over the print on any size paper, but the ads don't go into detail on fonts it can read, or state it can pull data from sources like newspaper clippings or magazine articles. (The price is $1250 qty 1, if anyone cares.) I recently got an ad from Palantir on their line and they claim their systems can read thousands of fonts, and they show a newspaper clipping in the ad's illustration of possible input. But it also looks like they are machines which want sheets fed in, so how would you input text from a book or newspaper? (I DON'T want to have to photocopy eveything first onto standard-sized paper!) Anyone have any advice on this subject? Regards, Will Martin "wmartin@almsa-1.arpa" -------
roddyprep@deneb.ucdavis.edu (0000;0000090825;420;755;401;) (07/19/88)
In article <12415307818.26.WMARTIN@SIMTEL20> WMartin@SIMTEL20.ARPA (William G. Martin) writes: >paper, but the ads don't go into detail on fonts it can read, or state >it can pull data from sources like newspaper clippings or magazine >articles. (The price is $1250 qty 1, if anyone cares.) Handheld scanners are rather popular in Japan for use with Japanese language word processors. They are in the $300 range as accessories to specific machines. Judging by the references in my NEC Word Processor manual, these units treat scanned text as a "picture" -- as bit-mapped graphics. You can manipulate the picture (rotate, crop, reverse, etc.) but you can not edit it as text. >I recently got an ad from Palantir on their line and they claim their >systems can read thousands of fonts, and they show a newspaper clipping >in the ad's illustration of possible input. But it also looks like they >are machines which want sheets fed in, so how would you input text from a >book or newspaper? (I DON'T want to have to photocopy eveything first onto >standard-sized paper!) Apparently these units were designed by engineers from the same school of design as those who did the first Xerox machines. Their philosophy - nobody but a dweeb what want to use anything other than 8.5 x 11 paper. One of my colleagues reports high error rates on reads even with clean copy. Moreover, the software on the Palantir lacks easy provision for "training" to get consistent misreads (mm -> rm, for example) to translate properly. Scanning is an area where the Japanese are likely to make substantial progress simply because once you've come up with a scanner for even printed Japanese everything else is simple. However, until they do, the choices are pretty grim unless you have 8.5 x 11 sheet copy in a clear, simple typeface....
schultz@mmm.UUCP (John C Schultz) (07/23/88)
In article <12415307818.26.WMARTIN@SIMTEL20> WMartin@SIMTEL20.ARPA (William G. Martin) writes: >Since the subject of scanners came up, I'd like to see anyone on the list >who has had experience with hand-held scanning units, or has done lots >of input from random published sources, like books, newspapers, and magazines, >post their comments and opinions. > My wife (a librarian) has some experience with hand-held and more expensive page scanners. Both types performed OCR internally, spitting out a stream of ACSII which had to be stored in a PC or something. She can't recall the name of the hand-held unit but the page scanner was a Kurzwiel (sp?) which is quite expensive. She has promised me some more info which I will pass on. In the meantime, her opinion was that the hand-held unit was not usable. The hand-held unit was very sensitive to the scan speed and the precision of the motion across the page. It also had a limited number of fonts it could recognize. The page scanner looked like a FAX machine and cost quite a bit. It also had a limited number of fonts but apparently did a much more reliable job with the fonts it could recognize. You will also need some program on your computer which will talk effectively to the scanner - check whether the manufacturer supplies this and how well things like Lotus, Word or other editors work. You will have to edit the scannerr input into usable format e.g. database. Hope this helps. -- john c. schultz schultz@mmm.UUCP (612) 733-4047 3M Center, Bldg 518-1-1, St. Paul, MN 55144-1000 The opinions expressed herein are, as always, my own and not 3M's.