Survival Analysis & Risk Models — Cox, RSF, DeepSurv — Hands-on

Learn how to model time to event outcomes in biomedical and omics studies. This module covers censoring, Kaplan–Meier curves, Cox models, random survival forests and DeepSurv style deep learning, with a focus on hazard ratios, discrimination, calibration and clinically meaningful risk score reporting in R and Python.

Survival Analysis & Risk Models — Cox, RSF, DeepSurv

Help Desk · WhatsApp

Session Index

Session 1 — Survival Data & Censoring Foundations Session 2 — Cox Proportional Hazards & Extensions Session 3 — Machine Learning Survival — RSF & Metrics Session 4 — DeepSurv, Risk Scores & Reporting

Session 1

Fee: Rs 8800

Survival Data & Censoring Foundations

Survival data structures and notation

time to event outcomes event indicators and censoring flag right censoring intuition

Censoring types and assumptions

administrative vs loss to follow up independent censoring idea truncation overview

Non parametric survival curves

Kaplan–Meier estimation median survival and confidence intervals log rank tests and group comparisons

Session 2

Fee: Rs 11800

Cox Proportional Hazards & Extensions

Cox proportional hazards model

hazard vs survival functions hazard ratios and interpretation partial likelihood idea

Model building and diagnostics

choice of covariates and transformations Schoenfeld residuals and PH checks influential observations

Extensions and special cases

time dependent covariates stratified Cox models competing risks overview

Session 3

Fee: Rs 14800

Machine Learning Survival — RSF & Metrics

Random survival forests (RSF)

ensemble idea for censored data survival trees and splitting rules variable importance concepts

Survival performance metrics

concordance index (C index) time dependent ROC and AUC integrated Brier score

Cross validation for survival models

patient level vs record level splits nested resampling concepts hyperparameter tuning for RSF

Session 4

Fee: Rs 18800

DeepSurv, Risk Scores & Reporting

DeepSurv style neural survival models

link to Cox proportional hazards loss network architecture and outputs regularization and overfitting control

Risk score calculation and stratification

linear predictors and risk groups calibration plots for survival decision curve analysis concepts

Deliverables: survival model and report pack

KM and adjusted survival curves C index, Brier and calibration summary R / Python scripts and model object

PDF

Survival Analysis & Risk Models — Cox, RSF, DeepSurv Training | Biostatistics & ML for Omics

Survival Analysis & Risk Models — Cox, RSF, DeepSurv — Hands-on