[Mrtrix-discussion] multiple bvalue

[Thijs Dhollander]

Hi Romain,

You should process them separately, because MRtrix (and CSD in general) expects single-shell data (apart from the optional appearance of one or more b=0 images for normalization): i.e. everything with a non-zero b-value will also be interpreted as originating from a single shell.  This would lead to erroneous results if you feed it data from multiple shells.

For example:
The data I typically work with, comes from a multi-shell acquisition that is optimized for DKI: 10x b=0, 25x b=700, 40x b=1000, 75x b=2800.  All this is done interleaved; so even though I can perfectly import the data in MRtrix (straight from Siemens dicoms: works perfectly and imports the gradient/b-value information without a problem in the correct frame of reference), I still have to sort out each single shell separately for CSD processing.

So, conclusion: only single-shell data for CSD. :-)

Cheers,
Thijs

-----Original Message-----
From: mrtrix-discussion-bounces at public.nitrc.org [mailto:mrtrix-discussion-bounces at public.nitrc.org] On Behalf Of romain valabregue
Sent: Wednesday, November 14, 2012 12:45
To: mrtrix-discussion at public.nitrc.org
Subject: [Mrtrix-discussion] multiple bvalue

Dear all

I have a data set with 2 dti acquisitions one with a b value of 1000 and one with 3000.
Is it possible to concatenate the 2 acquisitions and process them with mrtrix.
Does it make sense to have 2 bvalue mixte or should I process them separately ?

Many thanks


Romain Valabregue

_______________________________________________
Mrtrix-discussion mailing list
Mrtrix-discussion at www.nitrc.org
http://www.nitrc.org/mailman/listinfo/mrtrix-discussion