Hi - with one group having one protocol and the other having another I
don't think you can safely distinguish between protocol changes and
"true" group changes.
Cheers.
On 24 Dec 2008, at 09:21, Erick Canales Rodriguez (Benito Menni CASM)
wrote:
> Hi Steve and Janssen
>
> Thanks for your comments,
>
> I understand your points, in general with one value, e.g., '1' for
> the images with a particular voxel size and '2' for the other it is
> possible to remove the effect due to different sampling schemes.
> However, in my particular case the data is not well matched across
> the sampling schemes. That is, one group was measured with the first
> sampling scheme and the other with the second one, so, in this case
> the linear model is not able to separate the effect due to different
> sampling schemes from the effect due to different groups. The fit
> line will pass exactly through the center of the points that
> correspond to each group, consequently the distance between the line
> and the center of the points of both groups will be equal to zero,
> reason why the null hypothesis (i.e., not difference between the
> groups) will be fulfilled. For this reason I was interested in
> accurate interpolation methods, however, as Steve has pointed, the
> interpolation gives rise to bias that can not be modellable in a
> direct way.
>
> Any suggestion will be very useful for me,
>
> Thanks again and Best Regards,
>
> Erick J. Canales-Rodríguez
>
> (+34) 93 652 99 99 ext. 255
>
> Unitat de Recerca, Benito Menni CASM (http://www.hospitalbenitomenni.org/Default.aspx
> )
> Sant Boi del Llobregat (Barcelona), Espanya/España/Spain
>
> Benito Menni Complex assistencial en Salut Mental
> ________________________________________
>
>
>
> De: FSL - FMRIB's Software Library [[log in to unmask]] En nombre de
> Steve Smith [[log in to unmask]]
> Enviado el: martes, 23 de diciembre de 2008 18:51
> Para: [log in to unmask]
> Asunto: Re: [FSL] TBSS with different diffusion data?
>
> Hi I would just try (e.g.) '1' for one voxel size and '2' for the
> other, and then demean that EV. Cheers
>
> --------------------
> Stephen M. Smith, Professor of Biomedical Engineering
> Associate Director, Oxford University FMRIB Centre
>
> FMRIB, JR Hospital, Headington,
> Oxford. OX3 9 DU, UK
> +44 (0) 1865 222726 (fax 222717)
> [log in to unmask]
> http://www.fmrib.ox.ac.uk/~steve
> ----------------------
>
> On 23 Dec 2008, at 13:41, "Erick Canales Rodriguez (Benito Menni
> CASM)" <[log in to unmask]> wrote:
>
>> Hi Steve,
>>
>> Thanks for your reply,
>> In your opinion the spatial resolution of the data can be considered
>> as a covariable in the general linear model?
>> For example, one additional column (EV) to include the volume of the
>> voxel.
>>
>> Best regards,
>>
>> Erick J. Canales-Rodríguez
>>
>> (+34) 93 652 99 99 ext. 255
>>
>> Unitat de Recerca, Benito Menni CASM (http://www.hospitalbenitomenni.org/Default.aspx
>> )
>> Sant Boi del Llobregat (Barcelona), Espanya/España/Spain
>>
>> Benito Menni Complex assistencial en Salut Mental
>> ________________________________________
>> De: FSL - FMRIB's Software Library [[log in to unmask]] En nombre de
>> Steve Smith [[log in to unmask]]
>> Enviado el: martes, 23 de diciembre de 2008 14:02
>> Para: [log in to unmask]
>> Asunto: Re: [FSL] TBSS with different diffusion data?
>>
>> Hi,
>>
>> See previous posts on this: in general people are very wary of trying
>> to combine data across different protocols. You will have a bias in
>> pretty much all diffusion-derived measures, which may not be trivally
>> modellable, and is almost certainly not solvable by re-interpolating
>> your data, sadly. However, IF you put in the protocol labelling as a
>> confound regressor to randomise, and IF your regressors of interest
>> do
>> not correlate strongly with this, and IF your data is well matched
>> across the protocol groups, then you might get away with this........
>>
>> Cheers.
>>
>>
>>
>> On 23 Dec 2008, at 10:33, Erick Canales Rodriguez (Benito Menni CASM)
>> wrote:
>>
>>> Hello FSL,
>>>
>>> My interest is to compare two groups of subjects with TBSS, however,
>>> the diffusion MRI data corresponding to each group was recorded with
>>> a different spatial resolution (in the same MRI machine). That is,
>>> the same b-value and diffusion gradient directions but different
>>> voxel size, the first group: 1.13x1.13x5 mm and the second one:
>>> 0.94x0.94x5 mm.
>>> My question is, it is possible to use TBSS to infer FA differences
>>> in groups with data at different spatial resolutions? Typically this
>>> procedure is not a good practice, however, I consider this
>>> possibility due to the data was measured with a resolution lower
>>> than the resolution of the template used in TBSS: 1x1x1 mm.
>>> A possible solution may be to interpolate the resulting FA images to
>>> the same spatial resolution prior to the TBSS analysis. A second
>>> variant may be to perform the interpolation directly on the measured
>>> diffusion data. There is any script in FSL to perform accurately
>>> these interpolations?
>>> Many thanks in advance,
>>>
>>> Erick J. Canales-Rodríguez
>>>
>>> (+34) 93 652 99 99 ext. 255
>>>
>>> Unitat de Recerca, Benito Menni CASM (http://www.hospitalbenitomenni.org/Default.aspx
>>> )
>>> Sant Boi del Llobregat (Barcelona), Espanya/España/Spain
>>>
>>> Benito Menni Complex assistencial en Salut Mental
>>> ________________________________________
>
> Este mensaje se dirige exclusivamente a su destinatario y puede
> contener información privilegiada o confidencial. Si no es vd. el
> destinatario indicado, queda notificado de que la utilización,
> divulgación y/o copia sin autorización está prohibida en virtud de
> la legislación vigente. Si ha recibido este mensaje por error, le
> rogamos que nos lo comunique inmediatamente por esta misma vía y
> proceda a su destrucción.
>
> Aquest missatge va dirigit exclusivament al seu destinatari i pot
> contenir informació privilegiada o confidencial. Si vosté no és el
> destinatari indicat, li notifiquem que la utilització, divulgació i/
> o còpia sense autorització està prohibida en virtud de la legislació
> vigent. Si ha rebut aquest missatge per error, li preguem que ens ho
> comuniqui el més aviat possible.
>
---------------------------------------------------------------------------
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
---------------------------------------------------------------------------
|