Anjali wrote:

> I would like more information about how exactly the time series
> values of voxels are mapped onto the nodes in SUMA. Since there
> are ~30,000 voxels and the functional data are not
> interpolated, does this mean that each voxel's time series gets
> mapped 6x onto the nodes that correspond to the XYZ
> coordinates? Does this mean that some time series values are
> redundant in the niml.dset files? If you could explain in more
> detail how the values contained in the niml.dset files are
> obtained, I would appreciate it.

Information on that is in the slides in suma.pdf
[afni.nimh.nih.gov]
[afni.nimh.nih.gov]

also see 3dVol2Surf -help
>
> Also, I am not sure what is meant by the fact that the surface
> dataset is "sparse" by default.
Meaning if a node does not get values mapped onto it it is not listed in the output dataset.

When I use ConvertDset, the
> output contains 196,002 rows all with intensity values in them.
Suggesting, perhaps that you had values on all the nodes.

> If I were to use the pad_to_node max_index, what values would
> be substituted with 0 since all rows seem to have values in
> niml.dset file.

All rows having values, does not mean that all nodes have entries.
Perhaps the slides in suma.pdf would help clarify matters.

>
> Thanks,
> Anjali
cheers,
ziad