Image to Text

This project received 18 bids from talented freelancers with an average bid price of $1681 USD.

Get free quotes for a project like this
Employer working
Project Budget
$750 - $1500 USD
Total Bids
Project Description

See below that briefly describes what we are trying to accomplish. For any terminology, refer to the dictionary links towards the end of this email. Call me if you have any questions or would like to discuss this in detail. [Posting of direct contact details is strictly prohibited and will get your account suspended -
[url removed, login to view] Admin]

Within healthcare there is a system called PACS. This system stores reports and images that are generated from various medical equipment such as an ECHO machine. The images being stored in PACS are grouped together in DICOM files that are really large holding anywhere from 50-300 images in one DICOM file. One of these images is an actually report or measurements of the images within the file (see attached image) that we need to extract into text and map it over to a database. All the measurements in the image are required fields that need to be mapped over.

1) We need to figure out a way to pull this particular type of an image out of the DICOM file. There is no indication as to what image would contain this information or a file type we can look at to find these measurements, but we do know for sure that there can be 1 or more images containing all of the measurements we need and they are all within the DICOM file.

2) Once this image is found, it needs to extracted into text and mapped over to a database to populate the required fields. This usually happens by creating a transformation code between XML Tags.

1) There are OCR SDKs out in the market that can convert images to text. This is only usefully once we have figured out a way to pull the image we need from the DICOM file. Once the image is found, the SDK can convert it and then figure out a way to map it over to the database.

From my understanding the OCR SDK is able to convert the image into text / XML or XSLT that we need to map the fields.
Any code would work as long as it can solve the problem and work with the SDK being use.

PACS - [url removed, login to view]
ECHO - [url removed, login to view]
DICOM - [url removed, login to view]
OCR: [url removed, login to view]
SDK: [url removed, login to view]
XSLT: [url removed, login to view]

Looking to make some money?

  • Set your budget and the timeframe
  • Outline your proposal
  • Get paid for your work

Hire Freelancers who also bid on this project

    • Forbes
    • The New York Times
    • Time
    • Wall Street Journal
    • Times Online