Combining LSTM and Latent Topic Modeling for Mortality Prediction

TL;DR

This paper introduces a joint neural network model combining LSTM and latent topic modeling to improve mortality prediction accuracy from clinical notes, emphasizing interpretability and outperforming previous LDA-based models.

Contribution

It presents a novel end-to-end neural architecture that integrates LSTM with a constrained topic model for mortality prediction and interpretability.

Findings

Models outperform prior LDA-based approaches in mortality prediction.

The topic modeling layer is designed as a single-layer network with LDA-inspired constraints.

Limited success in interpreting topics from neural network weights.

Abstract

There is a great need for technologies that can predict the mortality of patients in intensive care units with both high accuracy and accountability. We present joint end-to-end neural network architectures that combine long short-term memory (LSTM) and a latent topic model to simultaneously train a classifier for mortality prediction and learn latent topics indicative of mortality from textual clinical notes. For topic interpretability, the topic modeling layer has been carefully designed as a single-layer network with constraints inspired by LDA. Experiments on the MIMIC-III dataset show that our models significantly outperform prior models that are based on LDA topics in mortality prediction. However, we achieve limited success with our method for interpreting topics from the trained models by looking at the neural network weights.