5ttedit to combine lesion volume with 5tt image volume

1

Hello everyone,
I tried to combine segmented lesion mask with 5tt image in order to perform global tractography.
I used this command

*5ttedit –path segmentedlesion.mif 5ttimage.mif 5tt_lesion.mif

But I encountered the problem with mismatch of two images dimensions. The error message was:
Image segmenedlesion_mask.mif does not match 5TT image dimensions
Then, I checked the size of images which is displayed below, and resized size of lesion file using:

*mrresize –size 142,156,142 input –interp linear output

mrinfo 5tt_nocoreg.mif

Image: “5tt_nocoreg.mif”

Dimensions: 142 x 156 x 142 x 5
Voxel size: 1 x 1 x 1 x ?
Data strides: [ 2 3 4 1 ]
Format: MRtrix
Data type: 32 bit float (little endian)
Intensity scaling: offset = 0, multiplier = 1
Transform: 0.9921 0.03527 -0.1206 -67.01
-0.01916 0.991 0.1322 -68.2
0.1242 -0.1288 0.9839 -58.05

mrinfo segmentedlesion.nii

Image: “segmentedlesion.nii”

Dimensions: 192 x 456 x 512
Voxel size: 1 x 0.488281 x 0.488281
Data strides: [ 1 2 3 ]
Format: NIfTI-1.1
Data type: unsigned 8 bit integer
Intensity scaling: offset = 0, multiplier = 1
Transform: 0.9921 0.03527 -0.1206 -81.84
-0.01916 0.991 0.1322 -92.29
0.1242 -0.1288 0.9839 -136.5
comments: TE=3.9e+02;Time=92512.870;phase=1

mrinfo segmentedlesion_mask_resize_linear.mif

Image: “segmentedlesion_mask_resize_linear.mif”

Dimensions: 142 x 156 x 142
Voxel size: 1.35211 x 1.42728 x 1.76056
Data strides: [ 1 2 3 ]
Format: MRtrix
Data type: 32 bit float (little endian)
Intensity scaling: offset = 0, multiplier = 1
Transform: 0.9921 0.03527 -0.1206 -81.73
-0.01916 0.991 0.1322 -91.75
0.1242 -0.1288 0.9839 -135.9

Then,I ran 5ttedit to merge lesion with 5tt image. In order to check if the lesion has been combined correctly, I ran 5tt2vis. Surprisingly, the position of lesion was somewhere else. As you can see in image, the green region is the lesion overlaid on 5tt image and in second image the crosshair is on the lesion incorporated in 5ttimage with value of 2 (default value for pathological tissue in 5tt2vis). The question is, why does 5ttedit change the image position while it has the same dimension as 5ttimage.

screenshot0003
screenshot0003613×731 124 KB
!

screenshot0004
screenshot0004613×731 124 KB

One more question, mrresize changed voxel size in addition to dimensions, does it make any kind of problems ?

2

I tried to combine segmented lesion mask with 5tt image in order to perform global tractography.

I hope by “global tractography”, you mean whole-brain tractography? “Global tractography” is a whole different beast.

mrresize –size 142,156,142 input –interp linear output
why does 5ttedit change the image position while it has the same dimension as 5ttimage.
One more question, mrresize changed voxel size in addition to dimensions, does it make any kind of problems ?

Yup. What’s happening here is that mrresize is changing the size of the voxels in order for the image FoV to contain the number of voxels that you have requested, while the actual size of the image FoV remains (approximately) unchanged. So while this gives you an image that contains the same number of voxels as the target image, it has not actually defined the image contents on the same image grid as the target image.

What you want to use instead is mrtransform.