Fancy that!!
Great news
T
Tim Behrens
Centre for Functional MRI of the Brain
The John Radcliffe Hospital
Headley Way Oxford OX3 9DU
Oxford University
Work 01865 222782
Mobile 07980 884537
[log in to unmask]
www.fmrib.ox.ac.uk/~behrens
On 10 Jul 2006, at 04:08, Susan Ilic wrote:
> Hi Tim,
> Quite amazing - flipping the X directions has made it looks really
> similar to your results now! Thanks a lot for your help.
> Susan.
>
>
>
>> ----- Original Message -----
>> From: "Susan Ilic" <[log in to unmask]>
>> To: [log in to unmask]
>> Subject: Re: [FSL] suggestions for improvement of dti data
>> Date: Tue, 4 Jul 2006 20:22:35 -0500
>>
>>
>> Hi Tim,
>> Thanks for that! I will give it a try. Funny that I'd noticed the
>> "lines" image did look a bit strange, but none of us picked up the
>> problem!
>> Regards,
>> Susan
>>
>>
>>> ----- Original Message -----
>>> From: "Tim Behrens" <[log in to unmask]>
>>> To: [log in to unmask]
>>> Subject: Re: [FSL] suggestions for improvement of dti data
>>> Date: Mon, 3 Jul 2006 20:07:38 +0100
>>>
>>>
>>> Hi Susan -
>>> these data look great, and I am sure you will be able to
>>> replicate the thalamus result with these.
>>>
>>> The problem you have is that you have the x direction (or
>>> equivalently the y and z directions) defined negative in your
>>> bvecs file.
>>>
>>> If you load dti_V1 into fslview in "lines" mode, you will easily
>>> be able to see this by looking at the corpus callosum in the
>>> coronal plane or axial plane. The lines don't join up!!!
>>>
>>> If you change the sign of all of your x entries in bvecs, and
>>> rerun dtifit/bedpost etc. all will be well
>>>
>>> Cheers
>>>
>>> T
>>>
>>>
>>> Tim Behrens
>>> Centre for Functional MRI of the Brain
>>> The John Radcliffe Hospital
>>> Headley Way Oxford OX3 9DU
>>> Oxford University
>>> Work 01865 222782
>>> Mobile 07980 884537
>>>
>>> [log in to unmask]
>>> www.fmrib.ox.ac.uk/~behrens
>>>
>>>
>>>
>>>
>>>
>>> On 3 Jul 2006, at 07:26, Susan Ilic wrote:
>>>
>>>> Hi Tim,
>>>> The ID number is 407232.
>>>> Thanks,
>>>> Susan
>>>>
>>>>
>>>>
>>>>> ----- Original Message -----
>>>>> From: "Tim Behrens" <[log in to unmask]>
>>>>> To: [log in to unmask]
>>>>> Subject: Re: [FSL] suggestions for improvement of dti data
>>>>> Date: Sat, 1 Jul 2006 07:32:07 +0100
>>>>>
>>>>>
>>>>> Hi -
>>>>> If you upload an input data directory to
>>>>> www.fmrib.ox.ac.uk/cgi-bin/upload.cgi
>>>>> and send me the relevant ID
>>>>> I'll take a look at it when I get the chance.
>>>>>
>>>>>
>>>>> Cheers
>>>>>
>>>>> T
>>>>>
>>>>>
>>>>>
>>>>> Tim Behrens
>>>>> Centre for Functional MRI of the Brain
>>>>> The John Radcliffe Hospital
>>>>> Headley Way Oxford OX3 9DU
>>>>> Oxford University
>>>>> Work 01865 222782
>>>>> Mobile 07980 884537
>>>>>
>>>>> [log in to unmask]
>>>>> www.fmrib.ox.ac.uk/~behrens
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On 30 Jun 2006, at 15:38, Susan Kouloyan-Ilic wrote:
>>>>>
>>>>>> Hi,
>>>>>> We have been trying to replicate your work on parcellating
>>>>>> the thalamus
>>>>>> using connectivity-based classification, to see if our data
>>> are up to the
>>>>>> challenge. Appears not....
>>>>>> The data have been acquired on a 3T GE machine, with 56
>>>>>> directions
>>>>>> (b=1200), 4 b0 images (b=7.9475). When we examine the tracts
>>>>>> from various
>>>>>> seed points all over the thalamus, they tend to form tight
>>>>>> loops around
>>>>>> the thalamus, basal ganglia, corpus callosum and surrounding
>>>>>> white matter,
>>>>>> as well as a tendency to go through the lateral ventricles.
>>>>>> Comparatively
>>>>>> very few reach the cortex. The results of the classification
>>>>>> also look
>>>>>> strange - instead of horizontal bands in the axial section,
>>>>>> we get some
>>>>>> lateral banding, as well as a predominance of frontal and
>>>>>> parietal
>>>>>> classification in the remainder.
>>>>>> We were lucky enough to have Mark Jenkinson visiting this
>>> week during the
>>>>>> fMRI experience, who examined our data. Apart from slightly
>>>>>> better
>>>>>> signal, and slightly more frontal lobe and brainstem >>>
>>> distortion, it looked
>>>>>> quite similar to the dataset he had available. He thought we
>>>>>> should
>>>>>> contact you for suggestions, and your opinion as to whether
>>>>>> cardiac gating
>>>>>> would be a critical part of data acquisition. (We did not
>>>>>> use gating).
>>>>>> He was kind enough to give us a copy of the data set used in
>>>>>> the recent
>>>>>> FSL course so we will try running our data with the masks to
>>>>>> try and
>>>>>> exclude that as a cause.
>>>>>> We'd love to hear some suggestions as to what might be the
>>>>>> problem and
>>>>>> what to do about it.
>>>>>> Thanks,
>>>>>> Susan Kouloyan-Ilic
>>>>>> Howard Florey Institute, Melbourne
>>>>
>>>>>
>>>>
>>>>
>>>> -- ___________________________________________________
>>>> Play 100s of games for FREE! http://games.mail.com/
>>
>>>
>>
>>
>> --
>> ___________________________________________________
>> Play 100s of games for FREE! http://games.mail.com/
>
>>
>
>
> --
> ___________________________________________________
> Play 100s of games for FREE! http://games.mail.com/
|