anonymization

100 views
Skip to first unread message

Miklos Espak

unread,
Nov 22, 2012, 10:03:27 AM11/22/12
to xnat_di...@googlegroups.com
Hi there,

I have a simple question about the site-wide / project specific anonymization.

Let's say I set up Kevin's script at site level.
(http://nrg.wustl.edu/projects/DICOM/BasicAppLevelConf.das)

If I create a project, on the 'Manage' tab I can enable or disable the
anonymization script. What happens if I enable it but leave the text
box empty? Will the data be anonymized with the site-wide script?

Will the project-wide script be launched if I disable the
anonymization at the project level?

Moreover, if I enable the anonymization at project level and add some
commands to the text box then is it gonna be executed 'after' or
'instead of' the site-wide script?

Thank you,
Miklos

Archie, Kevin

unread,
Nov 27, 2012, 11:00:32 AM11/27/12
to xnat_di...@googlegroups.com
Miklos,

The site-wide script is always applied, regardless of any other settings. If a project script is enabled, that is applied after the site script.

If you enable the project script but leave the text box empty, an exception gets thrown so no data can be archived. (Oops. Fixed in development code, should be available at next release. In the meantime, don't do that.)

- Kevin
--
You received this message because you are subscribed to the Google Groups "xnat_discussion" group.
To post to this group, send email to xnat_di...@googlegroups.com.
To unsubscribe from this group, send email to xnat_discussi...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/xnat_discussion?hl=en.


________________________________

The material in this message is private and may contain Protected Healthcare Information (PHI). If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.

Miklos Espak

unread,
Nov 27, 2012, 11:05:28 AM11/27/12
to xnat_di...@googlegroups.com
Thanks, Kevin, that's clear now.

Cheers,
Miklos

Jordi

unread,
May 17, 2013, 6:15:09 AM5/17/13
to xnat_di...@googlegroups.com
Hi,

I re-use this post since my doubt is very similar to the one initially posted by Miklos. 

Since I find it a pretty common decision, maybe some other users have had the same problem or know how to solve/workaround it.

I want my XNAT instance to have no site-wide script applied, instead I would like to define per project which anonymization script is applied (if any) to project's session data. Nevertheless, as long as I have experienced so far I cannot disable the site-wide script via web interface and all I can do is erase its content but then archival process will probably fail and no data would be archived. 

Any idea on how to achieve that?


thanks,
Jordi

Herrick, Rick

unread,
May 17, 2013, 3:16:50 PM5/17/13
to xnat_di...@googlegroups.com

Wait for the 1.6.2 release?

 

I’m kidding, but this is a bug that we found and fixed for 1.6.2, I believe. You can clear the site-wide anon script with a simple curl command, something like this:

 

curl --user admin:admin --request POST --data-ascii "anonScript=%2F%2F%20empty%20site-wide%20anon%20script" http://yourserver/data/services/settings

 

admin:admin of course becomes one of your system’s admin users.

 

One note: if you’re on Windows, you’ll need to double the % characters to escape them from the command shell interpreter, so %2F becomes %%2F, etc.

 

Actually, on re-reading your post, I’m not sure if I’m addressing exactly what you’re asking about. If you clear the anon script, this shouldn’t affect the archive process. It’s just that no anonymization will be performed based on the site-wide anon script. Any project- based scripts would still be applied. The bug I was addressing was a problem with updating and clearing the site-wide anon script. You are correct that we don’t have a feature to allow disabling the site-wide script. I believe this was a conscious decision, but I wasn’t actually part of that discussion so I’m not sure what the rationale was there.

To unsubscribe from this group and stop receiving emails from it, send an email to xnat_discussi...@googlegroups.com.


To post to this group, send email to xnat_di...@googlegroups.com.

Archie, Kevin

unread,
May 17, 2013, 3:29:58 PM5/17/13
to xnat_di...@googlegroups.com

I think there was a bug where an empty anon script created havoc – as Rick said, fixed in 1.6.2.

 

I would guess that if you can’t set an empty script, it would be possible to set a short script that includes just one operation that’s more or less certain not to affect a real DICOM file, maybe like:

 

-(0001,0f00)

 

I haven’t tested this, but it really seems like it should work.

 

-          Kevin

Herrick, Rick

unread,
May 17, 2013, 3:31:29 PM5/17/13
to xnat_di...@googlegroups.com

That curl command I had in my last post actually just set the script to be a comment:

 

// empty anon script

 

Or something like that. I don’t know if that would evaluate out to the same as an empty anon script though. It works in the dev tip code, but I haven’t tested it on 1.6.1.

Archie, Kevin

unread,
May 17, 2013, 3:34:02 PM5/17/13
to xnat_di...@googlegroups.com

I seem to remember that the bug involved scripts that didn’t include any operations – it wasn’t an empty script that caused problems so much as an empty parse tree.

 

Rick’s trivial script is better than mine if it works, but I suspect it won’t. Try his first and then mine.

Timothy Olsen

unread,
May 17, 2013, 6:46:35 PM5/17/13
to xnat_di...@googlegroups.com

I think you should be able to disable the site wide script entirely by doing a PUT to HOST/data/config/edit/image/dicom/status?activate=false

 

You might give that a shot first.

 

Tim

Jordi

unread,
May 23, 2013, 10:07:45 AM5/23/13
to xnat_di...@googlegroups.com
It does work, many thanks to all! 

One last thing, on what depends that the following script statement replaces the PatientName (0010,0010) DICOM tag value with either XNAT's subjectName/Label or subjectID?

(0010,0010) := subject

Jordi
Reply all
Reply to author
Forward
0 new messages