Deep Normed Embeddings for Patient Representation

TL;DR

This paper presents a new contrastive learning approach to embed clinical time series data into a geometric space where health states and risks are encoded as vector properties, aiding patient monitoring and decision-making.

Contribution

It introduces a novel geometric contrastive learning scheme for clinical data, linking vector norms to mortality risk and angles to organ failures, enhancing interpretability and utility.

Findings

Embedding encodes mortality risk via Euclidean norm.

Angles between vectors relate to organ system failures.

Improved downstream ML task performance and policy quality.

Abstract

We introduce a novel contrastive representation learning objective and a training scheme for clinical time series. Specifically, we project high dimensional EHR. data to a closed unit ball of low dimension, encoding geometric priors so that the origin represents an idealized perfect health state and the Euclidean norm is associated with the patient's mortality risk. Moreover, using septic patients as an example, we show how we could learn to associate the angle between two vectors with the different organ system failures, thereby, learning a compact representation which is indicative of both mortality risk and specific organ failure. We show how the learned embedding can be used for online patient monitoring, can supplement clinicians and improve performance of downstream machine learning tasks. This work was partially motivated from the desire and the need to introduce a systematic way of defining intermediate rewards for Reinforcement Learning in critical care medicine. Hence, we also show how such a design in terms of the learned embedding can result in qualitatively different policies and value distributions, as compared with using only terminal rewards.