Predicting local and distant metastasis for breast cancer patients using the Bayesian neural network

This paper presents a predictive accuracy comparison between the multivariate logistic regression (MLR) and the Bayesian neural network (BNN). The latter is presented in this paper as an alternative to the MLR (MLR). The MLR and BNN have been used to identify early breast cancer patients with high risk of tumour recurrence at the time of initial resection.