Fig. 1From: Predicting life expectancy with a long short-term memory recurrent neural network using electronic medical recordsSimplified LSTM architecture. At final time step t, xt represents the feature vector used as input to the hidden LSTM units, which activate output ht. In each preceding time step, output h functions as an intermediate prediction of life expectancy. We are interested in final prediction ht: a probability distribution for the next 50 monthsBack to article page