To calculate the survival curve choose the patient profile from the left panel controls and click 'Submit'.

To add a new patient profile redefine the characteristics and click 'Submit' again.

The application will compare the survivals for the selected patient profiles.

To display estimated survivals superimpose the slider on the curve

The marginal effect plots have been repredented for the four leading predictors according to the mimimal Depth of a variable measure. The prediction tool achieves an overall performance of 80.22% (C Index) on the survival outcome

Random Forest survival Predictions

Random Forests (RF) for Survival, Regression, and Classification (RF-SRC) is an ensemble tree method. Constructing ensembles from base learners such as trees can significantly improve learning performance. It was shown by Breiman[1] that ensemble learning can be further improved by injecting randomization into the base learning process --- a method called RF. RF-SRC (Random Forests for Survival, Regression, and Classification) extends Breiman's Random Forests method and provides a unified treatment of the methodology for models including right censored survival data [2].

The RF-SRC survival probabilities according to days, may be calculated defining in the right side panel the specific levels of predictors.

Computations have been performed using the randomForestSRC[1] package in R[3] (version 3.6.2).


1. Breiman L (2001). Random Forests. Machine Learning, 45(1), 5-32.

2. Ishwaran, H., Kogalur, U. B., Blackstone, E. H., & Lauer, M. S. (2008). Random survival forests. The annals of applied statistics, 2(3), 841-860.

3. Team, R. C. (2015). R Foundation for Statistical Computing; Vienna, Austria: 2014. R: A language and environment for statistical computing, 2013.