I am new to MRtrix3 and I am having troubles accessing the voxel coordinates.
I have .mif file with FOD and all I can read from that are SH coefficients of each voxel. In MRtrix view I can see coordinates of each voxel of interest so the information seems to be there.
How do I generate the matrix with voxel index and coordinates ?
Ok, then I guess the question is how are you currently getting the fixel information in the format you need? Because you’d need the voxel position to figure out where to find the data for each fixel. Do the docs on the fixel format not provide the information you’re after?
The question is expressed unusually, but I think I might be able to translate.
Storing “a matrix of voxel indices and coordinates” would be a highly inefficient way to handle such spatial information. The image transform permits, for any individual voxel, derivation of the location of that voxel in space, based on the coordinates of that voxel within the image in the three spatial axes.
Just in case: If you happen to additionally be reading ODF information directly from the .mif file raw data without using any provided I/O interface, and therefore you don’t have those indices in the three spatial dimensions but just a “voxel index” as the raw data are arranged in the file, then you will need to wrap your head around strides, which define how a 1D array of values stored within a file is mapped to axes of a higher-dimensional image.
The original question was about coordinates for individual voxels, I think, i.e. to go with the SH coefficients (of the FOD) in each voxel? Connectome nodes are something completely different. I’ll leave the latter to others to answer, if that’s effectively what you’re after.
OK, I’m sorry if my question was irrelevant, I’m still very new to MRTrix.
I produced my connectome using tck2connectome command and now I want the coordinate of the ROI nodes that are in my connectome. I was wondering if you can help me get that information
For any future web travelers: If what you ultimately wanted was, for each unique index in a parcellation image, the geometric centroid of each parcel in 3D space, this will be made more accessible with the next update of MRtrix3 via a new command labelstats.