C-mix: a high dimensional mixture model for censored durations, with applications to genetic data

TL;DR

This paper introduces C-mix, a high-dimensional mixture model for censored durations, with efficient inference and application to genetic data, outperforming existing models in predictive accuracy.

Contribution

The paper develops a novel high-dimensional mixture model for censored durations with Elastic-Net penalization and an efficient QNEM algorithm, applied to genetic datasets.

Findings

C-mix outperforms CURE and Cox models in C-index and AUC(t)

Efficient QNEM algorithm with proven convergence

Effective sparse parameterization in high-dimensional settings

Abstract

We introduce a mixture model for censored durations (C-mix), and develop maximum likelihood inference for the joint estimation of the time distributions and latent regression parameters of the model. We consider a high-dimensional setting, with datasets containing a large number of biomedical covariates. We therefore penalize the negative log-likelihood by the Elastic-Net, which leads to a sparse parameterization of the model. Inference is achieved using an efficient Quasi-Newton Expectation Maximization (QNEM) algorithm, for which we provide convergence properties. We then propose a score by assessing the patients risk of early adverse event. The statistical performance of the method is examined on an extensive Monte Carlo simulation study, and finally illustrated on three genetic datasets with high-dimensional covariates. We show that our approach outperforms the state-of-the-art, namely both the CURE and Cox proportional hazards models for this task, both in terms of C-index and AUC(t).