Dear Michael,
>>
>> This is in relation to FUGUE in FSL. I was using FUGUE on a phantom
>> to determine whether or not more accurate results of the phantoms
>> ADC could be attained after distortion correction. The ADC map of
>> the phantom was calculated with DTIFIT before and after the
>> correction and the results were compared. Before distortion
>> correction an ADC of 1.1211±0.012 ×10-3 mm2/s was calculated and
>> after the distortion correction an ADC of 1.1005±0.014 ×10-3 mm2/s,
>> was calculated. However , I know the ADC of the phantom is probably
>> around 1.15.
>>
>> My question is, should FUGUE provide more accurate values of ADC
>> after distortion correction or is it just a qualitative tool? So in
>> other words does FUGUE correct signal intensity values in the
>> distortion correction?
this is a question with two relevant answers, so I'll give both
1. The geometric distortions in DTI image also leads to intensity
distortions (due to stretching/compression). These distortions are
multiplicative and common to all acquisitions (i.e. both b0 and dwi
images). Because they are common to all scans they will divide out
when calculating the ADC values, and shouldn't affect those. If you
are observing slightly different values after fugue correction it is
probably because of a combination of smoothing induced by the tri-
linear interpolation and the ROI you use to measure this value.
2. In principle any correction of geometric distortions should include
also a correction for the accompanying intensity distortion (see
above). However, the intensity correction is based on the derivatives
of the field and is hence a very noisy measure. Therefore the default
behaviour of fugue is not to apply this correction. You can however
make it perform it by specifying the --icorr option.
Jesper
|