Anonymizing Tractoflow Outputs

Software Support
1

Our group has worked and produced a dataset of Tractoflow derivatives we plan to distribute openly.
The analyses and the paper are complete and we are working on meeting the data sharing requirements.

We’ve been informed we need to de-identify the entire sample in order to share the data. Specifically, this means removing or replacing the subject IDs from everywhere they exists.

Is there an existing process for anonymizing data this way that works with Tractoflows output structure? My main concern is not with the file names themselves, but with metadata that may have captured this information in the header. Is there any tool or documentation for cleaning and verifying this kind of process on imaging and tractography data?

Thanks for any help.

2

Hello Brent @bcmcpher,

We don’t have such a tool but I’ll take some time during the weekend to think about what kind of metadata could have embedded the original subject ID. I’ll get you an answer beginning of next week.

Best,
Arnaud

3

Hello @bcmcpher ,

I quickly checked and no metadata store subject ID. You should be okay with just renaming files.

Best,
Arnaud

4

Awesome, thanks. That’s what I’d thought from my own checks, but I wanted to verify if there was a standard process in place before I looked into making my own.