Deep Survival Analysis

TL;DR

This paper introduces deep survival analysis, a hierarchical generative model for EHR data that jointly models observations conditioned on latent structures and aligns data by failure time, improving risk stratification.

Contribution

It presents a novel deep hierarchical generative approach to survival analysis that handles heterogeneous data and aligns observations by failure time, outperforming traditional methods.

Findings

Deep survival analysis outperforms Framingham risk score in stratifying CHD risk.

The model effectively handles large-scale, heterogeneous EHR data.

It provides a scalable, joint modeling framework for survival analysis.

Abstract

The electronic health record (EHR) provides an unprecedented opportunity to build actionable tools to support physicians at the point of care. In this paper, we investigate survival analysis in the context of EHR data. We introduce deep survival analysis, a hierarchical generative approach to survival analysis. It departs from previous approaches in two primary ways: (1) all observations, including covariates, are modeled jointly conditioned on a rich latent structure; and (2) the observations are aligned by their failure time, rather than by an arbitrary time zero as in traditional survival analysis. Further, it (3) scalably handles heterogeneous (continuous and discrete) data types that occur in the EHR. We validate deep survival analysis model by stratifying patients according to risk of developing coronary heart disease (CHD). Specifically, we study a dataset of 313,000 patients corresponding to 5.5 million months of observations. When compared to the clinically validated Framingham CHD risk score, deep survival analysis is significantly superior in stratifying patients according to their risk.