TraCeR: Transformer-Based Competing Risk Analysis with Longitudinal Covariates

TL;DR

TraCeR is a transformer-based survival analysis framework that effectively incorporates longitudinal covariates and assesses calibration, leading to significant performance improvements over existing methods.

Contribution

It introduces a novel transformer-based model for survival analysis that handles longitudinal data and evaluates calibration, addressing key limitations of prior models.

Findings

TraCeR outperforms state-of-the-art methods on multiple datasets.

The model captures temporal covariate interactions without data assumptions.

It provides a comprehensive evaluation including calibration metrics.

Abstract

Survival analysis is a critical tool for modeling time-to-event data. Recent deep learning-based models have reduced various modeling assumptions including proportional hazard and linearity. However, a persistent challenge remains in incorporating longitudinal covariates, with prior work largely focusing on cross-sectional features, and in assessing calibration of these models, with research primarily focusing on discrimination during evaluation. We introduce TraCeR, a transformer-based survival analysis framework for incorporating longitudinal covariates. Based on a factorized self-attention architecture, TraCeR estimates the hazard function from a sequence of measurements, naturally capturing temporal covariate interactions without assumptions about the underlying data-generating process. The framework is inherently designed to handle censored data and competing events. Experiments on multiple real-world datasets demonstrate that TraCeR achieves substantial and statistically significant performance improvements over state-of-the-art methods. Furthermore, our evaluation extends beyond discrimination metrics and assesses model calibration, addressing a key oversight in literature.