Adversarial Time-to-Event Modeling

TL;DR

This paper introduces an adversarial deep learning approach for nonparametric estimation of event-time distributions in survival analysis, effectively handling censored data and outperforming traditional models.

Contribution

It presents a novel adversarial network framework for survival analysis that estimates event-time distributions without assuming parametric forms, incorporating censored data effectively.

Findings

Significant performance improvements over parametric models.

Effective handling of censored events in survival analysis.

Validated on both benchmark and real datasets.

Abstract

Modern health data science applications leverage abundant molecular and electronic health data, providing opportunities for machine learning to build statistical models to support clinical practice. Time-to-event analysis, also called survival analysis, stands as one of the most representative examples of such statistical models. We present a deep-network-based approach that leverages adversarial learning to address a key challenge in modern time-to-event modeling: nonparametric estimation of event-time distributions. We also introduce a principled cost function to exploit information from censored events (events that occur subsequent to the observation window). Unlike most time-to-event models, we focus on the estimation of time-to-event distributions, rather than time ordering. We validate our model on both benchmark and real datasets, demonstrating that the proposed formulation yields significant performance gains relative to a parametric alternative, which we also propose.