2.3.2.1 First-level feature

The first-level feature consists in computing the first order statistics of voxels within each VOI. The nth subject can be represented as:

$$r\_n^m = \left[r\_{n1}^m, r\_{n2}^m, \dots, r\_{np}^m\right]$$

$$\mathbf{r}\_n^s = \left[r\_{n1}^s, r\_{n2}^s, \dots, r\_{np}^s\right]$$

where r<sup>m</sup> <sup>n</sup> and r<sup>s</sup> <sup>n</sup> are the mean intensity and standard deviation of each VOI, respectively, and p is the number of VOIs; here p = 116.
