Survival analysis of censored data: neural network analysis detection of complex interactions between variables

Abstract

Neural networks can be used as pattern recognition systems in complex data sets. We are exploring their utility in performing survival analysis to predict time to relapse or death. This technique has the potential to find easily some types of very complex interactions in data that would not be easily recognized by conventional statistical methods. In this paper we demonstrate that there are several ways neural networks can be used to find three-way interactions among variables. Thus, in data sets where such complex interactions exist, neural networks may find utility in detecting such interactions and in helping to produce predictive models.