Hi,
The displacement needs to use the rotation as well as the translation.
The calculation in rmsdiff is more complicated than you've written, and is
described in the technical report at:
http://www.fmrib.ox.ac.uk/analysis/techrep/tr99mj1/tr99mj1/index.html
All the best,
Mark
> Hi,
>
> I have a problem when extracting head motion.
> I can get the head motion displacement in two ways:
> 1. with rmsdiff directly;
> 2. with avscale to get the Translations(x,y,z) firstly and calculate the
> length of the vector;
>
> There seems some difference between the results of these two methods.
> As I understand, rmsdiff calculate the root mean square with the formula
> sqrt((x*x+y*y+z*z)/3) while I calculate the displacement with
> sqrt((x*x+y*y+z*z)).
> However, this would not be the reason.
>
> Do you know that is the difference between these two methods?
>
> Best!
>
> Xiangzhen
>
>
>
>
> ¡¾Ð¡¿×³ÉÏñ¡¿http://www.conxz.net
>
> From: Stephen Smith
> Date: 2012-10-16 00:55
> To: FSL
> Subject: Re: [FSL] Get head motion parameters from DTI data?
> Hi
>
>
> On 15 Oct 2012, at 10:08, Xiangzhen Kong wrote:
>> Hi, Smith.
>> Thanks very much.
>>
>> Do you mean that eddy_correct is just some flirt processes?
>> (I have checked the script of eddy_correct, it seems that it is in that
>> case.)
>
> well then - you know the answer - yes!
>
>
>> Now my understanding is:
>> Through the edited eddy_correct in which -omat is following FLIRT, I can
>> get a transformational matrix from each weighted volume to the b0
>> volume.
>> Then I can get N (# of weighted images) matrix-files.
>> And then, I can get the displacements (deltaX, deltaY, deltaZ) for each
>> volume from these matrix-files.
>
> correct - you can get mean displacements via rmsdiff.
>
> Cheers.
>
>
>>
>> do you think so?
>>
>> Best!
>>
>> Xiangzhen
>>
>>
>> Xiangzhen Kong
>> http://www.conxz.net
>>
>> From: Stephen Smith
>> Date: 2012-10-12 14:15
>> To: FSL
>> Subject: Re: [FSL] Get head motion parameters from DTI data?
>> Hi - if you edit eddy_correct you can get FLIRT to save the transforms,
>> which is what you want.
>> Cheers.
>>
>>
>>
>> On 9 Oct 2012, at 15:51, Xiangzhen Kong wrote:
>>
>>> Hi, all.
>>> I want to extract some head motion parameters from raw DTI data? Do you
>>> have some idea about how to get them?
>>> And what do you think about these parameters? Do they have significance
>>> for understanding humans?
>>>
>>> Best!
>>>
>>> Xiangzhen Kong
>>
>>
>> ---------------------------------------------------------------------------
>> 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
>> ---------------------------------------------------------------------------
>>
>>
>>
>>
>
>
> ---------------------------------------------------------------------------
> 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
> ---------------------------------------------------------------------------
|