Estimation of Postmortem Period by Means of Artificial Neural Networks

The issue of estimating the postmortem period has always been a serious problem. Current methods do not provide satisfactory solutions. The problem is highly nonlinear and the variables involved are many and various. In this work we aim to propose a new method for estimating the postmortem period. This method is based on artificial neural networks. We use Multilayer Feedforward Networks. Learning takes place in supervised mode. We give a comparative study on a sample of 257 individuals to prove the advantage brought by this new technique, improving in this way the precision of the estimates given by the traditional methods.