DeepCENT: Prediction of Censored Event Time via Deep Learning

TL;DR

DeepCENT is a novel deep learning method designed to directly predict individual event times, effectively handling competing risks and outperforming existing approaches in survival analysis.

Contribution

It introduces a new deep learning framework with an innovative loss function for direct event time prediction, including handling competing risks.

Findings

DeepCENT outperforms traditional methods in simulation studies.

It effectively predicts individual event times in cancer datasets.

Handles competing risks in survival analysis.

Abstract

With the rapid advances of deep learning, many computational methods have been developed to analyze nonlinear and complex right censored data via deep learning approaches. However, the majority of the methods focus on predicting survival function or hazard function rather than predicting a single valued time to an event. In this paper, we propose a novel method, DeepCENT, to directly predict the individual time to an event. It utilizes the deep learning framework with an innovative loss function that combines the mean square error and the concordance index. Most importantly, DeepCENT can handle competing risks, where one type of event precludes the other types of events from being observed. The validity and advantage of DeepCENT were evaluated using simulation studies and illustrated with three publicly available cancer data sets.