Compensating trajectory bias for unsupervised patient stratification using adversarial recurrent neural networks

TL;DR

This paper identifies a trajectory bias in unsupervised patient stratification using RNN autoencoders and proposes an adversarial training method to mitigate this bias, improving the clinical relevance of patient embeddings.

Contribution

The paper introduces an adversarial training scheme to reduce trajectory bias in RNN autoencoder-based patient embeddings, enhancing unsupervised stratification.

Findings

The trajectory bias significantly affects patient clustering results.

The proposed adversarial method effectively reduces trajectory bias.

Improved clinical relevance of patient embeddings demonstrated.

Abstract

Electronic healthcare records are an important source of information which can be used in patient stratification to discover novel disease phenotypes. However, they can be challenging to work with as data is often sparse and irregularly sampled. One approach to solve these limitations is learning dense embeddings that represent individual patient trajectories using a recurrent neural network autoencoder (RNN-AE). This process can be susceptible to unwanted data biases. We show that patient embeddings and clusters using previously proposed RNN-AE models might be impacted by a trajectory bias, meaning that results are dominated by the amount of data contained in each patients trajectory, instead of clinically relevant details. We investigate this bias on 2 datasets (from different hospitals) and 2 disease areas as well as using different parts of the patient trajectory. Our results using 2 previously published baseline methods indicate a particularly strong bias in case of an event-to-end trajectory. We present a method that can overcome this issue using an adversarial training scheme on top of a RNN-AE. Our results show that our approach can reduce the trajectory bias in all cases.