Skip to main content
. 2012 May 19;34(10):2635–2654. doi: 10.1002/hbm.22092

Table 5.

Parameters used for the ANIMAL algorithm for human nonlinear transformations estimated for subcortical segmentation

Step size (mm) Iterations Gaussian blur (mm) Feature type
4 20 None Intensity
2 20 None Intensity
1 15 None Intensity

All values for lattice diameter were set to 3 x Step_size. Regularization parameters of stiffness, weight, and similarity were set to 1, 1, and 0.3, respectively as per the optimization of (Robbins et al., 2004).