Hi all,
I'm trying to use avwstats to count the number of voxels and find the mean
within a masked image. What I've done is created a binary mask, then
multiplied that by an image to only leave values in the non-masked regions.
So, to count voxels I'm typing:
avwstats <input> -V
It outputs 2 values; I understand that the 1st value is the number of voxels
that are non-zero. I've been getting whole numbers for this on the order of
hundreds (e.g., 826), which is what I expect. However, I don't understand
what the 2nd value is. I've been getting numbers on the order of thousands,
with decimals for this (e.g., 13560.75). Can someone explain what this
number is?
To find the mean, I'm typing:
avwstats <input> -M
It outputs a value that I believe is the mean for all the non-zero voxels.
However, I realized that it's possible that some of the volumes will have
zero values within the non-masked regions. So, is there some way to mask
regions so that the masked area is converted to NaNs - or some other method
that would allow me to include zeros when calculating the mean?
Thanks,
Pat
|