Causal predictive modeling of survival of Breast cancer: A comparative study with machine learning, traditional survival modeling methods along and causal survival models

Principal Investigator

Name
Malinda Iluppangama

Degrees
PhD

Institution
Loyola University Maryland

Position Title
tenure-track Assistant Professor

Email
miluppangama@loyola.edu

About this CDAS Project

Study
PLCO (Learn more about this study)

Project ID
PLCO-2009

Initial CDAS Request Approval
Jan 8, 2026

Title
Causal predictive modeling of survival of Breast cancer: A comparative study with machine learning, traditional survival modeling methods along and causal survival models

Summary
This investigation delves into the potential application of data-driven survival modeling approaches for prognostic assessments of Breast cancer survival. The primary objective is to evaluate and compare the ability of machine learning (ML) models and conventional survival analysis techniques to identify consistent key predictors of cancer survival outcomes. This study employs real data-driven survival modeling approaches to predict cancer survival, including survival-specific methods such as the Cox Proportional Hazards (CPH) model, Random Survival Forests (RSF), and Cox Proportional Deep Neural Networks (DeepSurv) and transformer-based survival models. Furthermore, we will utilize causal inference and causal survival random forest to estimate and predict the average treatment effect of early-stage diagnosis on Breast cancer patient survival.

Aims

The primary objectives of our study can be defined in the following manner,

We want to develop data-driven survival models by utilizing traditional, machine learning, and transformer based model to understand the survival of patients with breast cancer.

Compare the performance of the proposed models utilising well-defined methods and identify the association between risk factors and the survival time.

Increase the interpretability of the survival models by utilizing the Shapley Additive Explanation (SHAP) method

Finally, we want to conduct causal inference and a causal survival random forest to estimate and predict the average treatment effect of early-stage diagnosis on Breast cancer patient survival.

Collaborators

Malinda Iluppangama Loyola Maryland
Dilmi Abeywardana Loyola Maryland
Hansapani Rodrigo University of Texas Rio Grande