Hello
I have a bunch of data that I analyzed with fmriprep 1.3.2. I haven’t had any issues with it and for the sake of simplicity would like to use it going forward. Are there any bugs in fmrirep 1.3.2 that could be messing up my data that I should know about?
Hi Noam,
The main thing to be aware of is a bug that was fixed in 1.5.10/20.0.6, but that affected all previous versions, including 1.3.2. To quote the changelog:
This release fixes a bug for phase-difference fieldmaps that are not in RAS+ orientation. The bug presented as an error if the orientation was reordered relative to RAS+ (for example, AIL+) and the swapped dimensions were not of the same size. Otherwise, the bug introduced a poor masking of the phase difference map, and could be quite subtle if the original orientation was LAS+.
I would have a look through the changelog and search for “FIX” to see if there are other issues that might affect your data, but that was the main one I can recall where data might be silently misinterpreted.
As usual, our general recommendation is to use the more recent versions for new studies, but to continue using the same version for an ongoing study as long as it continues to work for your data. We did release 20.2 as a long-term support series that will continue to be maintained with only bug fixes for 4 years. If you are starting a new study, I would encourage you to consider upgrading to that.
Thank you for the reply. Guess I will rerun it, my fieldmaps were collected in RPI.