Document Management - page level indexing

Community Topic(s):

Keywords: document management, page, Indexing, remittance, EOB, healthcare

By ,

March 01, 2012 - 11:42 AM

I have been looking for a document management solution that will give me page level indexing, with multiple sets of indexes per page.  The solutions I am finding are all document level indexing. 

Here is an example of what I am looking for:  A 50 page healthcare document with 200 patients listed in that document.  I would like to index each patient to the page they are located on, also there could be mulitple patients on a single page and I need the ability to link all of those patients to the page they are on.  The retrieval process should allow a patient account number to be entered and take me to the page that patient's data is on.

I do NOT want a solution where you have to search twice, such as search for the document then do a search of an OCR'd document to find the patient account number. 

Let me know if any of you have found a solution for this.

Thanks.

Report

Add a Response

You need to log in to post messages. Click here to login.

Responses

Submitted by

I would look at full text indexing for the entire document. Then you should be able to get a hit list of every page in your search scope that had the patient id you searched on.

Report
Was this helpful? Yes No
Reply

Submitted by

Thank you for the response Scott.
I am looking for a solution that does not use OCR technology but rather manual indexing capabilities at the page level. I've gone down the full-text search path before using multiple OCR voting engines (tried with and without OCR zones) with little success. What we found was that the time spent in OCR and correction took more time per page than indexing directly from the page, especially when you are only needing a few key values from the page. This was entirely due to the quality of the images we received, but we have no control over that. Our images come from bank lockboxes and the quality is poor. The images are human readable but the OCR engines could never seem to pick up the data effectively enough to use that technology.

Report
Was this helpful? Yes No
Reply

Submitted by

Kurt I see how that would be a challenge with poor quality originals for OCR. You could look at a couple different document management applications that I have worked with over the years. Westbrook technologies has one called Fortis and EMC Application extender. In either of those I have been able to index line item detail for invoices at the page level. Different data but similar concept. When I used those they each had the ability to have multilevel fields that would expand based on the number of line items on the invoice and then became searchable with the queries. Are you using a system today?

Report
Was this helpful? Yes No
Reply

Submitted by

Kurt, Sorry I just looked at an old Virtual Machine and tried this with Fortis it still only gets you to the document level, so I misspoke that's what I get for trying to rely on my memory from that long ago.

Report
Was this helpful? Yes No
Reply

Submitted by

I appreciate the help. We currently use Imagenow from Perceptive Software but they also do document level only indexing. Their solution was to slice the documents into little pieces and then make each slice a document. While that would give me what I need, it does not maintain the integrity of the document which is also very important.

Let me know if you see or hear of a solution.

Thanks again.

Report
Was this helpful? Yes No
Reply

Submitted by

Hi Kurt, If you are still searching for a soltion perhaps you could contact chris.driver@docstoruk.com
I spoke with him about your question and he believes he has a solution.
My apologies if this is not the usual protocol in responding to Community Topics!

Report
Was this helpful? Yes No
Reply

Submitted by

We do this currently in our FileStore EDM software by using PDF files and then creating at page level bookmarks which can hold free text of say one or more patient names.

We can also annotate specific words on the page and again tage these as well.

The metadata can be assigned to the annotations and bookmarks either manually or in bulk from other metadata.

When searching our search interface can then search across and within documents for these bookmarks or annotations.

Our software could be purchased as a full solution or as OEM/Toolkit componenets to integrate with other products.

Let me know if you wish to discuss further.

Regards

Kevin Ingram
www.dcs.co.uk

Report
Was this helpful? Yes No
Reply

Submitted by

Kurt,we have a solution (Orienge) that can index to the page if you input attributes (electronically or manually as indicated to be your preference at this time). With further discussion to clarify salient details, we might propose an even better solution that could eliminate steps, save large blocks of time, and deliver results on all desired specs. I would be happy to discuss further.

Report
Was this helpful? Yes No
Reply

Submitted by

You can see www.finedocs.com. Product has feature for page level indexing.

Report
Was this helpful? Yes No
Reply
Blog email facebook linkedin Rss Twitter Youtube

Copyright © AIIM 2013 Privacy Policy