Radial Basis Function Neural Network for Classification of Quantitative EEG in Patients with Advanced Chronic Renal Failure - Foundations on Natural and Artificial Computation - page 414

Fig. 2. Scatter plot matrix and corresponding values of the Pearson coecient between
chosen parameters derived from qEEG
EEG derivation. The parameters are listed in table 2. So, 55 parameters were
obtained for each subject.
Using these characteristic features as the input set, a neural network classifier
using RBF is implemented as a custom made MATLAB script.
The net input to the radial basis transfer function is the vector distance
between its weight vector, w , and the input vector, p , multiplied by the bias, b .
The RBF has a maximum of one when its input is zero. As the distance between
w and p decreases, the output increases. Thus, a radial basis neuron acts as a
detector, which produces one whenever the input, p , is identical to its weight
vector, w . A probabilistic neural network, a variant of radial basis network, was
used for the classifications. When an input is presented, the first layer computes
distances from the input vector to the training input vectors and produces a
vector whose element indicates how close the input is to a training input. The
second layer sums these contributions for each class of inputs to produce net
output vector probabilities. Finally, a complete transfer function on the output
of the second layer picks the maximum of these probabilities and produces a one
for that class and a zero for the other classes.