Maybe this link helps? Have a look at

http://plastimatch.org/dicom_comparison.html

Thanks, that is a very usefull link!

Dicom Cleaner : http://www.dclunie.com/pixelmed/software/webstart/DicomCleanerUsage.html

I am partial to the RSNA free offering - RSNA mirc dicom editor and anonymizer. mircwiki.rsna.org

You can try ORS Visual Lite. It is a free dicom viewer with anonymization function.

http://theobjects.com/en/support/downloadLite.php

http://doradiology.com/DICOManonymizer/index.html

Fantastic free app for Mac OS

DicomBrowser does good job at modifying DICOM tags. Recently used it to anonymize set of DICOM images that came from different x-ray machines, and to modify the tags so they appeared as being a part of one study.

http://nrg.wustl.edu/software/dicom-browser/

I second for DicomBrowser. It also has its own simple language (much simpler than MIRC anonymizer) in which you can write instructions for conditional anonymization and execute in batches.

The easiest App to use is: http://doradiology.com/DICOManonymizer/index.html

The developer kindly made a modified version for a Chest Radiologists (Teaching File) Collaboration:

http://chestradiologists.org/blog/2014/05/26/title/

Howard Mann

DicomBrowser cannot be downloaded from the site.

I was disappointed about the free offerings, since the ones I looked at really did not perform a good anonymization "out of the box".  Some of them were very configurable, but then you had to be very knowledgeable about what is actually considered identifiable personal information.

The configuration for anonymization is a big effort to take on, and a big responsibility to use a freeware without doing a thorough validation of all your possible DICOM attributes.  For instance, many manufacturers have dangerous private elements, some which are necessary to keep for different modalities.  Some modalities have sequences (tree structures of DICOM elements) which are not anonymized.

Because of this I have spend 3 years developing a networked hardware device (it is a DICOM reciever), to which you can transfer images and let it perform a full anonymization.  The good thing is that you do an export and anonymization from any work station within a few seconds, instead of the normally time consuming methods.  It also sorts images in a folder structure which is easy to understand (which is not what you get from typical DICOM exports).  It also has a web-interface for relaxing anonymization, and for validating with clear indications if there are any dangerous elements remaining (this is really only happening if you modify default settings).

Sorry, but here comes a section you may skip if your are sensitive to non-free software: I am now trying to sell this commercially under the name Dicom2usb (www.dicom-port.com).  I believe the cost is a lot less than putting in the effort to develop from scratch, especially when considering that Dicom2usb has rules (keep/modify/remove etc) for 3000 DICOM elements and a large number of private elements.

So I guess, apart from advertising my venture, I really want to tell you that it is a big effort to do anonymization thoroughly and that you may very well use free software.  But you should be very, very sure about what you are doing, and take a good anonymization for granted without validating it to the DICOM standard.

Recommended reading is the DICOM standard, especially the chapters about de-identification.

MATLAB has a function for anonymizing DICOM files

Please see:

http://www.mathworks.com/help/images/ref/dicomanon.html

http://www.mathworks.com/help/images/ref/dicomanon.html

I think that the best is the one from Pixelmed publishing:

http://www.dclunie.com/pixelmed/software/webstart/DicomCleanerUsage.html

This can be used in conjunction with other tool and has the unique feature to blackout images and series of them.

Actually the blackout is a feature that permits to really anonymize images, not only the contents of DICOM fields.

"Some types of images, particularly ultrasound, screen shots of analysis panels and 3D rendering, digitized film or paper documents, and some nuclear medicine images, contain identifying information not just in the DICOM "header", but also burned in to the pixel data, which needs to be "blacked out" (redacted) by replacing the pixel values using an image editor."

You have to  do this anonymization, for example  in Dose CT report, Ultrasound images, Angiography Secondary capture, and other types.

Also it has a the possibility to receive images from DICOM nodes and to send images to other DICOM Storage SCP.

It is very reliable and stable, and if you choose to go with webstart installation, it updates automatically itself every time there is a new  release.

If you are looking for a free tool to guarantee patient safety and privacy in your DICOM data records, I recommend the modiCAS||anonymizer.

Due to high flexibility, you can determine the degree of anonymization yourself, from the patient´s name to all essential data containing information about the patient and your institution.

The modiCAS||anonymizer is available free of charge.

Download: http://www.modicas.de/en/modicas-anaonymizer.html

Link Update for recommended modiCAS||anonymizer:

https://www.modicas.de/modicas-anonymizer-dicom-anonymisieren.html

Download: full Version, free of ads, free of charge.

If you want to convert you DICOMs to NIFTI anyway, you can also use dcm2nii:

http://people.cas.sc.edu/rorden/mricron/dcm2nii.html

Isn’t it impossible to anonymize radiological images? I thought it’s like with a finger print. Pseudonymization might be possible after stripping off the meta data from the DICOM file. But I’m not sure, maybe there are body parts of which the images truly can be anonymized. I’d be very interested if there are.

@LeonieLampe The inherent uniqueness of a person's anatomy doesn't preclude successful anonymization of their radiological images. For example, given a chest x-ray without any accompanying metadata, it would not be possible for a third party who does not already possess a fully identified copy of the exam to infer the identity of the patient from the image pixels of that chest x-ray. See https://www.hhs.gov/hipaa/for-professionals/privacy/special-topics/de-identification/index.html?language=es for HHS's guidance on the topic.

This is, in my opinion, the best free dicom anonymization tool:

http://santec.tudor.lu/project/dicom/download

for Mac OS Horos will do the job

Our free ProSurgical3D software includes DICOM anonymization:

https://www.stratovan.com/products/pro-surgical-3d?utm_source=rgps3d&utm_medium=ans&utm_campaign=anon1

Just load the scan, then on the Export tab choose the method of anonymization. A statement indicating how it is anonymizing the scan is also shown so you'll know what's happening.

http://medimodel.com/guide/how-to-anonymise-a-dicom-file/

medimodel can aid in dicom images anynomization

A Python resource:

https://github.com/KitwareMedical/dicom-anonymizer

And? Did you test all of them? Also, do these tools only takeout dicom labels or also anonimize eg the faces in the images themselves?

DicomAnonymizer is an open source app, very good and small code

https://sourceforge.net/projects/dicomanonymizer/

Margarethus M Paulides this platform doesn't provide features to have a proper discussion of the different answers. But I'm glad to have these suggestions compiled here for people to discover and try out for themselves.

https://github.com/Emory-HITI/Niffler

http://dicom.nema.org/medical/dicom/current/output/chtml/part04/chapter_Z.html

Dear;

Please check the following links :

https://github.com/Emory-HITI/Niffler

Regards

The best Dicom anonymizer is Dicom Cleaner:

http://www.dclunie.com/pixelmed/software/webstart/DicomCleanerUsage.html

From David Clunie.

Has anyone come across the diminished quality of the scans once they have been anonymized?

DICOM Cleaner

3D Slicer

MIPAV has a DICOM Anonymizer tool

3D slicer

We have listed a number of DICOM anonymization tools on 'imaging QA' our free medical imaging Question and Answer site. They can easily be listed using our search tool :

https://community.imagingqa.com/docs

DICOM library does a good job too.

DICOM+

https://github.com/irrer/DICOMClient#user-content-dicom-for

- GUI

- command line for bulk anonymization

- free

- anonymizes UIDs and retains consistent references between series. (e.g. struct and image)

- approved for use by US Dept. of Veterans Affairs

- previewer

- edit metadata

You can use the new handy anonymizer from IMV team. https://anonymizer.imv-ms.com/

This is one of the best DICOM anonymizers you could ever find. And it is available for free here:

https://www.ai-medical.ch/adage

you can use the convenient and free program imv anonymizer https://anonymizer.imv-ms.com/