Using the DICOM Pixel Anonymizer with very large DICOMs

[Brandon Driscoll]

Hi,

I've been using CTP client for ages but typically our plan to deal with burn in has been to filter out the images that typically contained it using the dicomfilter.

That's worked fine but now we are dealing with video fluoro studies which have burn in at the same place on every frame which should be easy to remedy with the pixel anonymizer.  However the image sets are also massive individual DICOMs as much as 1.2 GB.  They are single image multiframe. If I try and send them with 32 bit JAVA they just silently fail since it runs out of memory.  With 64 bit java they work without issue.  

When I tried out the pixelanonymizer I got a "rejected by dicomfilter" in the client window.  When I started digging I noticed that you can't use the pixel anonymizer with 64-bit java (though I'm not sure if that's the root of the error message).

So it seems I'm stuck between a rock and a hard place, either the files won't send or I can't use the pixel anonymizer.  Any suggestions?

The relevant parts of my scripts in the CTP client are below

In config.default

dfEnabled=yes

dfName=dicom-filter.script

dpaEnabled=yes

dpaScriptName=dicom-pixel-anonymizer.script

daScriptName=dicom-anonymizer.properties

showURL=no

showBrowseButton=yes

zip=yes

dicom-pixel-anonymizer.script

{ Modality.equals("RF") }

(0,0,100,20) (480,200,32,250)

Thanks,

Brandon Driscoll

Research Associate

QIPCM Team, Techna Institute

University Health Network

[John Perry]

Brandon:

 

If you get a "rejected by DicomFilter" message, there must be something in the image that doesn't match the DicomFilter script. The DicomFilter is called before the pixel anonymizer. When that message occurs, the pixel anonymizer isn't even called, so it isn't the source of the problem.

 

It would be interesting to compare the image to the script to see what test is failing. Normally, I would ask you to send me the image and your script, but if the image is that big, it would take me forever to download it (I live in a corn field, and my ISP is not much better than two tin cans and a string).

 

If you have a way to make an element listing of the image, try checking it against your DicomFilter script. If you don't have a utility that would do that, let me know and I'll point you to a couple of them.

 

You can run CTPClient on a 64-bit Java, but since the pixel anonymizer needs the ImageIO Tools, you might have a problem on a Windows system since the tools are only available in 32 bits. If you are running a 64-bit Java on a Windows system and your Java is Java 8 or older, you can put the two ImageIO Tools jars in the Java/jre/lib/ext directory. That will get you at least some codecs, but depending on the image in question, maybe not the one you need. See http://mircwiki.rsna.org/index.php?title=Java_Advanced_Imaging_ImageIO_Tools for more information. A zip file containing the jars is available at http://mirc.rsna.org/ImageIO/win-i686.zip

 

JP

--
You received this message because you are subscribed to the Google Groups "RSNA MIRC CTP/TFS User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rsnas-ctpmirc-user...@googlegroups.com.
To post to this group, send email to rsnas-ctpmir...@googlegroups.com.
Visit this group at https://groups.google.com/group/rsnas-ctpmirc-user-group.
For more options, visit https://groups.google.com/d/optout.

[Brandon Driscoll]

Sorry I misstyped. I meant rejected by dicompixelanonymizer.

> You received this message because you are subscribed to a topic in the Google Groups "RSNA MIRC CTP/TFS User Group" group.
> To unsubscribe from this topic, visit https://groups.google.com/d/topic/rsnas-ctpmirc-user-group/wLlfV9Q3uTw/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to rsnas-ctpmirc-user...@googlegroups.com.

[Brandon Driscoll]

I've done some more testing and found some other data Fluoro data sets where I don't get the rejected by dicompixelanonymizer message using either 32 or 64 bit Java so it seems to just be the set itself.

The image set is about 245 MB but I just sent a different 245 MB one from a different site and it wiped the correct part of the frame.

I think there may just be something about this one video.

If I can ever recreate it I'll send it across.

Thanks again,

Brandon

> To unsubscribe from this group and stop receiving emails from it, send an email to rsnas-ctpmirc-user-group+unsub...@googlegroups.com.
> To post to this group, send email to rsnas-ctpmirc-user-group@googlegroups.com.

> Visit this group at https://groups.google.com/group/rsnas-ctpmirc-user-group.
> For more options, visit https://groups.google.com/d/optout.
>
> --
> You received this message because you are subscribed to a topic in the Google Groups "RSNA MIRC CTP/TFS User Group" group.
> To unsubscribe from this topic, visit https://groups.google.com/d/topic/rsnas-ctpmirc-user-group/wLlfV9Q3uTw/unsubscribe.

> To unsubscribe from this group and all its topics, send an email to rsnas-ctpmirc-user-group+unsub...@googlegroups.com.
> To post to this group, send email to rsnas-ctpmirc-user-group@googlegroups.com.