Deep learning models for the analysis of high-dimensional survival data with time-varying covariates while handling missing data
Open Access
- 24 July 2025
- journal article
- research article
- Published by Springer Nature in Discover Artificial Intelligence
- Vol. 5 (1) , 1-17
Abstract
Recent advances in deep learning have expanded the potential for predictive modeling in survival analysis, particularly in high-dimensional datasets with time-varying covariates. This paper applies deep learning approaches, DeepSurv, DeepHit, and Dynamic DeepHit, to model HIV incidence (time-to-event outcome) using high-dimensional longitudinal data, incorporating time-varying cytokine profiles alongside baseline covariates. We employ the time-dependent concordance index (C-index) and Brier scores to assess the models’ predictive accuracy. We also address missing data using missForest, evaluating model performance on imputed and complete-case datasets. Different strategies for integrating cytokine profiles were explored: DeepSurv and DeepHit utilized derived variables, mean, and difference between the first and last measurements, while Dynamic DeepHit preserved the original time-varying nature of the cytokine data. Our findings demonstrate that retaining the dynamic nature of cytokine covariates, rather than relying on derived summary measures, underscores the robustness and suitability of Dynamic DeepHit as a clinical prediction model, particularly in scenarios where key variables evolve over time.Keywords
Funding Information
- Sub-Saharan Africa Consortium for Advanced Biostatistics (SSACAB II) programme (DEL-22-009)
This publication has 48 references indexed in Scilit:
- The role of deep learning‐based survival model in improving survival prediction of patients with glioblastomaCancer Medicine, 2021
- Parameter Tuning Using Adaptive Moment Estimation in Deep Learning Neural Networks2020
- Dynamic predictions with time‐dependent covariates in survival analysis using joint modeling and landmarkingBiometrical Journal, 2017
- Gradient methods for minimizing composite functionsMathematical Programming, 2012
- Estimating a time‐dependent concordance index for survival prediction models with covariate dependent censoringStatistics in Medicine, 2012
- Consistent Estimation of the Expected Brier Score in General Survival Models with Right‐Censored Event TimesBiometrical Journal, 2006
- A time‐dependent discrimination index for survival dataStatistics in Medicine, 2005
- Assessment and comparison of prognostic classification schemes for survival dataStatistics in Medicine, 1999
- A neural network model for survival dataStatistics in Medicine, 1995
- Multivariate survival analysis using Cox's regression modelHepatology, 1987