We use the 3dSurftoVol command in a few of our analysis pipelines - often using the different mapping functions for different data types. We were curious if anyone is able to provide insight into how the program would behave with multimodal data when using the ‘mode’ mapping function? i.e., Is one mode chosen or are they combined somehow?
That “mode” may not be what you are thinking of. The mode here refers to the most common value sampled along the line segment at each node connecting a pair of surfaces or along a normal vector. So if a segment contains values:
1 2 3 6 6 3 7 6
The mode is 6. The maximum, mean, median, minimum will all give different values of course. It might be useful to know how you want to project data from the surface to the volume.
Thanks for the response Daniel. Our question is what happens if instead of the values you provided, the values along the segment are 1 2 3 6 6 3 3 6. Now there are two mode values: 3 and 6. What value is mapped?
Not sure this matters as we assume the mode function works the same in both directions, but to be accurate we are actually asking about this in relation to 3dVol2Surf mapping rather than the other way around as written in our original post.
The modes are not combined. In both directions (v2s, s2v), the mode should come out as the smallest of the modes, 3 in your example.
That’s exactly what we were looking for - thanks for confirming Rick.