The Usefulness of Artificial Neural Networks in Predicting the Outcome of Hematopoietic Stem Cell Transplantation

Abstract

1.1 Artificial neural networks in clinical medicine In Medicine, several tools have been developed for the prediction of clinical outcomes following drug treatment and other medical interventions. The standard approach for a binary outcome is to use logistic regression (LR), however, this method requires formal training and a profound knowledge of statistics (Royston, 2000; Harrel et al., 1996). LR is used to predict a categorical (usually dichotomous) variable from e set of predictor variables; it has been especially popular with medical research in which the dependent variable is whether or not a patient has a disease. Over the past years, artificial neural networks (ANNs) have increasingly been used as an alternative to LR analysis for prognostic and diagnostic classification in clinical medicine (Schwarzer et al., 2000). ANNs are composed of simple elements operating in parallel inspired by biological nervous systems. As in nature, the network function is determined largely by the connections between elements. After training with retrospective data ANNs are capable of making intelligent predictions given new, limited information. The growing interest in ANNs has mainly been triggered by their ability to mimic the learning processes of the human brain. However, the issue remains as to how these ANNs actually succeed in recognizing patterns within data that are too complex for the human brain. From here derives the so-called “black-box” aspect of ANNs. The network operates in a feed-forward mode from the input layer through the hidden layers (like in a black box) to the output layer. Exactly what interactions are modeled in the hidden layers is still a knot that remains untied. Each layer within the network is made up of computing nodes with remarkable data processing abilities. Each node is connected to other nodes of a previous layer through adaptable inter-neuron connection strengths known as synaptic weights. ANNs are trained for specific applications, such as pattern recognition or data classification, through a learning process and knowledge is usually retained as a set of connection weights. The backpropagation algorithm and its variants are learning algorithms that are widely used in neural networks. With backpropagation, the input data is repeatedly presented to the