Deep conditional transformation models for survival analysis

TL;DR

This paper introduces deep conditional transformation models (DCTMs) that unify parametric and semiparametric survival analysis, effectively handling complex, non-tabular data and various censoring types, and demonstrating competitive performance with current deep learning methods.

Contribution

The paper proposes DCTMs as a flexible, unified framework for survival analysis that accommodates non-linear hazards and diverse data types, extending existing neural network approaches.

Findings

DCTMs perform competitively with state-of-the-art deep learning methods.

They effectively model non-linear and non-proportional hazards.

The approach handles all types of censoring and truncation.

Abstract

An every increasing number of clinical trials features a time-to-event outcome and records non-tabular patient data, such as magnetic resonance imaging or text data in the form of electronic health records. Recently, several neural-network based solutions have been proposed, some of which are binary classifiers. Parametric, distribution-free approaches which make full use of survival time and censoring status have not received much attention. We present deep conditional transformation models (DCTMs) for survival outcomes as a unifying approach to parametric and semiparametric survival analysis. DCTMs allow the specification of non-linear and non-proportional hazards for both tabular and non-tabular data and extend to all types of censoring and truncation. On real and semi-synthetic data, we show that DCTMs compete with state-of-the-art DL approaches to survival analysis.