What should be happening in the scenario where i've concatenated 6 scans together?

They were scanned back to back during the same scanning session, with the same 15 direction + 2 b0 protocol.  I concatenated them all together into a long 4D nii.gz ( including the bvalues and bvectors ).

I was expecting DTIPrep to average all the B0 and treat the all the gradient directions individually, but what actually happened is that they were averaged into a single 17 direction scan, then processed.  Also, my b0s were never averaged.

i should mention, this was on a Mac with version 1.2.4

same version on Linux with the same protocol/options behaved as i would've expected

Hi Chris

Yes, you should be able to concatenate, we do it all the time in our studies

Did you try to feed DTIPrep directly the nii.gz image? If so, I don't think that this would work, you would have to first convert the image to NRRD (with DWIConvert). 

Also Re b0, Currently DTIPrep does not allow to specify the minimal b-value for a b0 to be detected. Only those gradients with very small gradient entry (10-7) are currently detected as b0.
Something we need to improve.

Btw, for combining multiple DWI's you can also use CombineDWIs https://github.com/NIRALUser/CombineDWIs
Martin
Originally posted by Christopher Petty:

Hey Martin .. Just coming back to this ..

I did process them as nrrd ... the linux version behaved as i would've expected, it averaged all the B0s together for me ... but the same protocol processed on the mac had weird results i described above and It did not average the B0 and

mm, sounds like a mac related bug. Can you double check that both linux and mac versions are the same and if so at a bug report on the tracker?

Thanks
Martin