Deep Survival Analysis for Competing Risk Modeling with Functional Covariates and Missing Data Imputation
This work addresses prognostic modeling in critical care by handling dynamic risk factors and incomplete data, representing an incremental advance in domain-specific survival analysis.
The paper tackles survival analysis under competing risks with functional covariates and missing data by introducing the Functional Competing Risk Net (FCRN), which integrates functional data representation and imputation in an end-to-end deep-learning model, showing substantial improvements in prediction accuracy over baselines like random survival forests and traditional models in simulated and real-world ICU datasets.
We introduce the Functional Competing Risk Net (FCRN), a unified deep-learning framework for discrete-time survival analysis under competing risks, which seamlessly integrates functional covariates and handles missing data within an end-to-end model. By combining a micro-network Basis Layer for functional data representation with a gradient-based imputation module, FCRN simultaneously learns to impute missing values and predict event-specific hazards. Evaluated on multiple simulated datasets and a real-world ICU case study using the MIMIC-IV and Cleveland Clinic datasets, FCRN demonstrates substantial improvements in prediction accuracy over random survival forests and traditional competing risks models. This approach advances prognostic modeling in critical care by more effectively capturing dynamic risk factors and static predictors while accommodating irregular and incomplete data.