Figure 2 .

Structure of a radial basis function neural network adapted from González- Camacho et al. (2012). In the hidden layer, information from input variables $(x_{i1},\dots ,x_{ip})$ (j = 1,…,p markers) is first summarized by means of the Euclidean distance between each of the input vectors $\left\{x_i\right\}$ with respect to S (data-inferred) (k=1,…,S neurons) centers $\left\{c_k\right\}$, that is, $u_i^{[k]}=h_k\left|\right|x_i-c_k|{|}^2$. These distances are then transformed using the Gaussian function $z_i^{[k]}=\text{exp}(-u_i^{[k]})$. These scores are used in the output layer as basis functions for the linear regression $y_i=\mu +{\displaystyle \sum _{k=1}^S{w}_k{z}_i^{[k]}+{\epsilon }_i}$.