Learning and Naming Subgroups with Exceptional Survival Characteristics

TL;DR

This paper introduces Sysurv, a novel non-parametric method using random survival forests to identify and interpret subgroups with exceptional survival outcomes, overcoming limitations of existing approaches.

Contribution

Sysurv is a fully differentiable, non-parametric approach that automatically learns interpretable rules for subgroup identification based on individual survival curves.

Findings

Sysurv effectively identifies meaningful survival subgroups.

It outperforms existing methods on various datasets.

Case study demonstrates actionable insights in cancer data.

Abstract

In many applications, it is important to identify subpopulations that survive longer or shorter than the rest of the population. In medicine, for example, it allows determining which patients benefit from treatment, and in predictive maintenance, which components are more likely to fail. Existing methods for discovering subgroups with exceptional survival characteristics require restrictive assumptions about the survival model (e.g. proportional hazards), pre-discretized features, and, as they compare average statistics, tend to overlook individual deviations. In this paper, we propose Sysurv, a fully differentiable, non-parametric method that leverages random survival forests to learn individual survival curves, automatically learns conditions and how to combine these into inherently interpretable rules, so as to select subgroups with exceptional survival characteristics. Empirical evaluation on a wide range of datasets and settings, including a case study on cancer data, shows that Sysurv reveals insightful and actionable survival subgroups.