Skip to content
forked from chl8856/DeepHit

DeepHit: A Deep Learning Approach to Survival Analysis with Competing Risks

Notifications You must be signed in to change notification settings

texchi2/DeepHit

 
 

Repository files navigation

DeepHit

Title: "DeepHit: A Deep Learning Approach to Survival Analysis with Competing Risks"

Authors: Changhee Lee, William R. Zame, Jinsung Yoon, Mihaela van der Schaar

Description of the code

This code shows the modified implementation of DeepHit on Metabric (single risk) and Synthetic (competing risks) datasets.

The detailed modifications are as follows:

  • Hyper-parameter opimization using random search is implemented
  • Residual connections are removed
  • The definition of the time-dependent C-index is changed; please refer to T.A. Gerds et al, "Estimating a Time-Dependent Concordance Index for Survival Prediction Models with Covariate Dependent Censoring," Stat Med., 2013
  • Set "EVAL_TIMES" to a list of evaluation times of interest for optimizating the network with respect these evaluation times.

About

DeepHit: A Deep Learning Approach to Survival Analysis with Competing Risks

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%