Interpretable Fine-Gray Deep Survival Model for Competing Risks: Predicting Post-Discharge Foot Complications for Diabetic Patients in Ontario

TL;DR

This paper introduces CRISPNAM-FG, an interpretable deep survival model for competing risks that predicts diabetic foot complications post-discharge, balancing high accuracy with transparency for clinical trust.

Contribution

The paper presents a novel intrinsically interpretable survival model using Neural Additive Models with the Fine-Gray approach, enhancing transparency in competing risks prediction.

Findings

Achieves competitive predictive performance with existing deep survival models.

Provides transparent predictions through shape functions and feature importance plots.

Validated on multiple datasets including real-world hospital data.

Abstract

Model interpretability is crucial for establishing AI safety and clinician trust in medical applications for example, in survival modelling with competing risks. Recent deep learning models have attained very good predictive performance but their limited transparency, being black-box models, hinders their integration into clinical practice. To address this gap, we propose an intrinsically interpretable survival model called CRISPNAM-FG. Leveraging the structure of Neural Additive Models (NAMs) with separate projection vectors for each risk, our approach predicts the Cumulative Incidence Function using the Fine-Gray formulation, achieving high predictive power with intrinsically transparent and auditable predictions. We validated the model on several benchmark datasets and applied our model to predict future foot complications in diabetic patients across 29 Ontario hospitals (2016-2023). Our method achieves competitive performance compared to other deep survival models while providing transparency through shape functions and feature importance plots.