Thank you again for looking into this.
Motion parameters are within a degree in rotation and within 1 mm in
translation.. Even though acquisition was interleaved, when we use B-spline
interpolation the stripes in RPV.img are practically invisible for
normalization to EPI template. There are no stripes in ResMS image.
Please note that SPM default of trilinear interpolation also caused stripes in
RPV.img in images with isotropic voxels.
Thanks and Regards,
Sheeba
Quoting Torben Lund <[log in to unmask]>:
> Hi Sheeba
>
> That sounds in the normal range, are you sure it is not just motion,
> if you use interleaved acquisition it could look like this? Have a
> look at the ResMS image also, are the stripes also in this image?
>
>
> Best
> Torben
>
>
>
> On 20 Feb 2007, at 17:00, Sheeba Arnold wrote:
>
>> Thank you very much for trying to help me with this.
>>
>> The voxel size is 3x3x5.
>> We use trilinear interpolation (SPM default) for normalization..
>>
>> The stripes in RPV.img are practically invisible if I use higher
>> order B-Spline
>> interpolation for normalization to EPI template.
>>
>> Is there a recommended type of interpolation for non-isotropic voxels?
>>
>> Thanks and Regards,
>> Sheeba
>>
>>
>> Quoting [log in to unmask]:
>>
>>> Hi Sheeba
>>>
>>> What is your voxel size, are the voxels you use very long? I don't know
>>> how this would affect the RPV image in the extreme case, but first we need
>>> to know if your voxels are extreemly long.
>>>
>>> Best
>>> Torben
>>>
>>>
>>>> Thank you very much for the reply!
>>>>
>>>> The movement parameters look fine (within 1 mm, 1 degree)! What's
>>>> confusing is that we see stripes in RPV.img in all the subjects scanned
>>>> with this protocol.
>>>>
>>>> What information is contained in each voxel in RPV.img?
>>>>
>>>> The raw images look fine.
>>>>
>>>> In another study with isotropic voxels, we don't see such artifacts..
>>>>
>>>> Best Regards,
>>>> Sheeba
>>>>
>>>
>>>
>>>
>
>
|