Hi Mark,
Thanks for the explain.
Best,
Leo
On Wed, Sep 29, 2010 at 9:27 PM, vivian <[log in to unmask]> wrote:
> Thank you. I didn't answer this question since I was confused by the
> translation. Now I understand.
>
> On Thu, Sep 30, 2010 at 12:24 AM, Mark Jenkinson <[log in to unmask]>
> wrote:
>>
>> Hi,
>>
>> The answer is no.
>>
>> It is complicated by the fact that the centre of rotation is not at
>> (0,0,0)
>> and hence the translation component of the matrix is a combination
>> of the translations in the parameters and effectively induced translations
>> from the rotations. Best not to worry about this as you don't need to
>> know
>> in order to use the tools effectively.
>>
>> All the best,
>> Mark
>>
>>
>> On 29 Sep 2010, at 23:20, vivian wrote:
>>
>>> Hi Mark,
>>>
>>> May I ask if the line is Rx, Ry, Rz, Tx, Ty, Tz, should the last three
>>> components of the line be 0.533133, 0.187519, -2.275521 as given by the
>>> matrix?
>>>
>>>
>>> On Thu, Sep 30, 2010 at 12:16 AM, Mark Jenkinson <[log in to unmask]>
>>> wrote:
>>> Dear Leo,
>>>
>>> There is a mapping between the par and the mat files but it
>>> is not simple and unless you *really* need to know then I
>>> would advise not bothering with it.
>>>
>>> To regress out motion the simple thing to do is to use the
>>> parameters (from the .par file) directly as regressors.
>>> This is most easily done in FEAT using the GUI (there is
>>> a single button) but you can import the .par file as columns
>>> in a design matrix in any other GLM too.
>>>
>>> As to your last question, the parameters are:
>>> Rx Ry Rz Tx Ty Tz
>>> where R is a rotation (angle in radians) and T is
>>> a translation (in mm).
>>>
>>> All the best,
>>> Mark
>>>
>>>
>>>
>>> On 29 Sep 2010, at 17:59, Yiou Li wrote:
>>>
>>> Hi,
>>>
>>> I checked the history threads and Couldn't find the following question
>>> about McFlirt being addressed.
>>>
>>> Could any expert on McFlirt/Flirt kindly advise?
>>>
>>> Best,
>>> Leo
>>>
>>>
>>> ---------- Forwarded message ----------
>>> From: Yiou Li <[log in to unmask]>
>>> Date: Tue, Sep 28, 2010 at 3:42 PM
>>> Subject: Question about MCFLIRT output transformation
>>> To: FSL - FMRIB's Software Library <[log in to unmask]>
>>>
>>>
>>> Hi all,
>>>
>>> I performed a 6 DOF of mcflirt on a function MRI dataset and asked for
>>> output transformation as well as the parameters (i.e., -mats and
>>> -plots are both used).
>>>
>>> I checked the first line of the .par file:
>>>
>>> -0.0198375 -0.00176798 -0.0073474 -0.0507674 -0.458798 -0.553456
>>>
>>> and the first mat file (MAT_0000)
>>>
>>> 0.999971 -0.007347 0.001768 0.533133
>>> 0.007381 0.999776 -0.019836 0.187519
>>> -0.001622 0.019849 0.999802 -2.275521
>>> 0.000000 0.000000 0.000000 1.000000
>>>
>>> The questions are:
>>> 1. Is there is correpsondence between the lines of .par and the MAT
>>> data? If so, how is the mapping?
>>>
>>> 2. To regress out the motion effects, should the lines of .par be used
>>> and what are the specific meanings of each column? (e.g., rotation,
>>> translation etc.)
>>>
>>> Your input is appreciate!
>>> Leo
>>>
>>>
>>>
>>>
>>> --
>>> Xiaowei.Zou
>>>
>>> Department of Biomedical Engineering
>>> Columbia University
>>> New York, NY, 10027
>>>
>
>
>
> --
> Xiaowei.Zou
>
> Department of Biomedical Engineering
> Columbia University
> New York, NY, 10027
>
>
|