Hi - I suspect that the dropout is simply the normal dropout and
distortion that you get with the EPI imaging normally used in DTI and
FMRI. The normal approach to this is either to ignore it, or to apply
EPI unwarping by taking a fieldmap etc.
Cheers, Steve.
On 9 Apr 2007, at 22:28, Hedok Lee wrote:
> Dear Martin:
>
> Thanks for your response.
>> - firstly, the bias field correction is suppose to correct for the
>> B1 inhomogeneity
> Ok, I didn't know.
>> - and secondly, if you multiply the b=0 and your b=1250 images
>> with the same bias field correction map, the effect is discarted,
>> when the linear function is fitted to ln(S(b)/S(0))
>>
> Well, I should have explained it a little bit more detail on how I
> computed ADC and FA. What you mentioned is probably valid if you
> apply eddy current correction before modulating DTI images with the
> bias-field. Modulating the image intensity by the correction
> factor seems to change the amount of eddy correction, which in
> turn effect subsequent ADC and FA values.
>> If you see any intensity variations throughout your slices, first
>> make sure
>> your acquisition protocol is correct, namely your TR is
>> sufficiently long compare to the T1.
>>
> Thanks for the tips. I double checked TR/TE=5600msec/83msec.
> Isn't TR nearly 4~5 times the T1 of GM or WM in 3.0T? Again,
> correct me if I'm wrong here. You want to have long TR to achieve
> fully relaxed state(thermal equilibrium is the proper word?) isn't
> it? I thought having TR=5600msec is probably long enough.
>
> Thanks again,
>
> Hedok
>
>> Regards,
>>
>> Martin
>>
>> On Mon, 9 Apr 2007, Hedok Lee wrote:
>>
>>> Dear FSLers:
>>>
>>> This is regarding intensity variation in our DTI scan. We acquired
>>> 2.5x2.5x2.5mm(40 slices b=0,b=1250 12 directions) using Siemens 3.0T
>>> machine. After going through a FDT pipeline and compute ADC and
>>> FA maps,
>>> we noticed there is a small reduction in ADC near frontal lobe. The
>>> intensity of b=0 image also seems to suggest intensity reduction
>>> in the
>>> same area. I assumed this kind of intensity variation is due to
>>> bias-filed
>>> originated from inhomogenuity of a B0 field, and thus I tried to
>>> estimate
>>> the amount of bias-field by FAST. I segmented the b=0(T2W) image
>>> into 3
>>> compartments, and selected bias-field option. From what I
>>> understand the
>>> bias field image intensity represents a correction factor between I
>>> (observed)=I(ideal)*b(correction) as stated in the eq20 of the FAST
>>> technical note. To correct for the bias-field, I multiplied BOTH
>>> b=0 and
>>> b=1250 images by b(correction) image acquired from FAST, and
>>> reran the FDT
>>> pipeline again.
>>>
>>> I?m attaching ADC maps of before and after making the correction
>>> for the
>>> bias-field.
>>>
>>> I appreciate if someone can help me with this method. Is method
>>> valid?, ok
>>> but needs more work?, or definitely wrong?
>>>
>>> Any suggestion would be appreciated.
>>>
>>> Hedok
>>>
------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---
|