Use of radial basis functions in computer-aided diagnosis of prostate cancer

In this paper, we show the results of a study in which we try to test the feasibility of using radial basis functions neural networks (RBFs for short) in clinical decision support systems. We have implemented two instances of RBFs in order to diagnose possible prostate cancer cases from a clinical database. To give an idea about how good the results are, we follow a two-fold approach. On the one hand they are independently evaluated in terms of accuracy, sensitivity and specificity and on the other hand they are compared with the performance over the same database of a classifier widely applied to the medical field problems, as it is multi-layer perceptron (MLP). The experimental results show that RBFs are a useful tool to build up clinical decision support systems.