Dear Haakon
> First: I've yet to see any concrete recommendations as to optimal
> settings for B0 maps. I'm currently gathering field maps using a factory
> optimized sequence which leaves me with a phase and magnitude image that
> the Fieldmap toolbox accepts readily. However, the specific parameters of
> the field map, such as voxel size, and slice number is seldom in
> correspondence with the functional data I collect. Am I right in assuming
> that this is suboptimal, and that we SHOULD acquire field maps that are in
> exact correspondence to the functional images with regard to 1: voxel size
> 2: number of slices and 3: acquisition orientation. Am I forgetting
> something?
Assuming that you are acquiring gradient echo fieldmaps, the FieldMap
toolbox is designed to accept input for which 1) voxel size, 2) number of
slices and 3) acquisition orientation do NOT have to be the same as the
EPI. We standardly acquire fieldmaps of 3mm^3 voxel size regardless of the
EPI voxel resolution (not all of our EPI sequences have 3mm^3 voxels).
This voxel size is a good trade-off between acquisition time and the
spatial scale of field changes that can be corrected for.
Importantly, we recommend to acquire 64 transverse slices so that the
whole of the head is covered. This means that field information will be
acquired beyond the EPI slice coverage and avoids any unnecessary
truncation of the EPI data when the distortion correction is applied.
It is also important that the shim values/currents are the same for both
the EPI and the fieldmap data. On Siemens scanners it is possible to do
this by selecting 'Copy Parameters', from the EPI volume then selecting
'Adjust Volume' when prescribing the field map acquisition. This step
copies the information about the volume over which the shim is adjusted
for the EPI data into the prescription of the field map, ensuring that the
field was the same over this region. I'm not sure how this can be done
with Philips or GE scanners.
> Second: Suppose I have a rather large amount of data with fieldmaps
> for in which voxel size and number of slices is not in correspondence with
> the functional data. Is it still possible to perform fieldmap based
> unwarping?
Ideally, as described above, the field map should be acquired with the
same shim values as is acquired for the EPI data. The actual number of
slices acquired and voxel size is not important as the FieldMap toolbox
can take care of this using the positioning information encoded in the
image headers.
> Third: Suppose I run multiple experiments during one scanning session,
> in which the acquisition parameters(no. of slices, voxel size, slice gap,
> TR, TE) vary between experiment. Is it necessary to acquire a
> fieldmap matched to each experiment?
Again, the most important thing is the actual field information that is
acquired for the field map rather than the geometry.
Best wishes
Chloe
> I have a sneaking suspicion that the reason why I can't find these
> questions answered anywhere is because the answers are rather obvious to
> those well-versed in these matters, but I hope you will find time to
> answer me in spite of this.
>
> Best wishes,
> Haakon,
> University of Oslo
>
>
--
Dr Chloe Hutton
Wellcome Trust Centre for Neuroimaging
Institute of Neurology
University College London
12 Queen Square
London WC1N 3BG
United Kingdom
Tel. +44-20-7833-7478, (internal ext 4368)
Fax +44-20-7813-1420
[log in to unmask]
|