SurvivalGAN: Generating Time-to-Event Data for Survival Analysis

TL;DR

SurvivalGAN is a novel generative model designed to produce realistic survival data, effectively handling censoring and imbalance issues, thereby enhancing the quality of synthetic data for survival analysis tasks.

Contribution

The paper introduces SurvivalGAN, addressing specific challenges in generating survival data, including censoring and imbalance, with new metrics and mechanisms for improved synthetic data quality.

Findings

SurvivalGAN outperforms baselines in generating survival data.

It effectively mitigates failure modes related to censoring and at-risk imbalance.

Synthetic data improves downstream survival model performance.

Abstract

Synthetic data is becoming an increasingly promising technology, and successful applications can improve privacy, fairness, and data democratization. While there are many methods for generating synthetic tabular data, the task remains non-trivial and unexplored for specific scenarios. One such scenario is survival data. Here, the key difficulty is censoring: for some instances, we are not aware of the time of event, or if one even occurred. Imbalances in censoring and time horizons cause generative models to experience three new failure modes specific to survival analysis: (1) generating too few at-risk members; (2) generating too many at-risk members; and (3) censoring too early. We formalize these failure modes and provide three new generative metrics to quantify them. Following this, we propose SurvivalGAN, a generative model that handles survival data firstly by addressing the imbalance in the censoring and event horizons, and secondly by using a dedicated mechanism for approximating time-to-event/censoring. We evaluate this method via extensive experiments on medical datasets. SurvivalGAN outperforms multiple baselines at generating survival data, and in particular addresses the failure modes as measured by the new metrics, in addition to improving downstream performance of survival models trained on the synthetic data.