|
Abstract:
|
There has been increasing interest in modelling survival data using deep learning methods in medical research. Current approaches have focused on designing special cost functions to handle censored survival data. We propose a very different method with two steps. In the first step, we transform each subject's survival time into a series of pseudo conditional survival probabilities. In the second step, we use these pseudo probabilities as quantitative response variable in a deep neural network (DNN) model. By using the pseudo values, we reduce a complex survival analysis to a standard regression problem, which greatly simplifies the neural network construction. Furthermore, we investigate the corresponding Bayesian DNN model and using variational inference to conduct inference for various quantities of interest.
|
