This topic and its replies were posted before the current version of LEADTOOLS was released and may no longer be applicable.
#1
Posted
:
Monday, July 1, 2019 12:22:56 AM(UTC)
Groups: Registered
Posts: 2
In the demos it didn't look like DICOM was an available type to OCR. Would I have to convert to another type (lossless TIFF) and then OCR, then how to redact (blackout), burn and then get back to a DICOM.
Trying to pull info out and anonymize them. Or do I turn to TIFF OCR< get the coordinates and then annotate the DICOM. I want to make sure I keep the original resolution of the DICOM for machine learning purposes.
Any sample code super helpful.
THanks!
#2
Posted
:
Monday, July 1, 2019 7:53:11 AM(UTC)
Groups: Manager, Tech Support, Administrators
Posts: 218
Was thanked: 12 time(s) in 12 post(s)
Hello,
If your goal is to anonymize DICOM files, then we have a demo specifically for that functionality in the SDK:
C:\LEADTOOLS 20\Examples\DotNet\CS\DicomAnonymizerDemo
Demonstrates the DICOM Anonymizer Toolkit functionality.
- Anonymize any DICOM Data set
- Editor shows differences between original an anonymized values
- Can create, save, and load scripts to customized the anonymization process
Hadi Chami
Developer Support Manager
LEAD Technologies, Inc.
#3
Posted
:
Monday, July 1, 2019 8:58:23 AM(UTC)
Groups: Registered
Posts: 2
Thanks Hadi! I did see this demo. I already handled the tags. My big concern now is the clean the images themselves. I wasn't sure if the OCR engine could natively read a DICOM. I'd like to find -any- characters a-z, A-Z or 0-9 and just redact them to black. Wasn't sure how to do it to preserve the quality of the original DICOM. Convert to lossless TIFF (or other format?) then OCR and find the coordinates of text and then use the DICOM annotation tool to permanently burn? Not sure if you've done it or thought through it before. Thanks.
#4
Posted
:
Monday, July 1, 2019 9:28:22 AM(UTC)
Groups: Manager, Tech Support, Administrators
Posts: 218
Was thanked: 12 time(s) in 12 post(s)
The OCR Engine can read in DICOM files as images natively, you wouldn't have to convert them to another format first. I am still not understanding the use-case for this though. Do your DICOM files have text burned into the images already? Typically, the DICOM files contain the tags and the pixel element (image) then on rendering in a viewer, the text is rendered on top of the image data. So if you anonymize the tags, the viewer should pull the anonomous tags instead of the original.
Can you send us a sample DICOM file that you are working with so that I can see what you are referring to? I sent you an email from
support@leadtools.com - please reply with the DICOM file attached so that I can see what you are referring to.
Thanks,
Hadi
Hadi Chami
Developer Support Manager
LEAD Technologies, Inc.
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.