A Transformer-based survival model for prediction of all-cause mortality in heart failure patients: a multi-cohort study

2025-03-16

AI Summary

To address insufficient accuracy and generalizability in long-term mortality risk prediction for heart failure (HF) patients, we propose TRisk—the first Transformer-based survival prediction model for longitudinal electronic health records (EHR). TRisk is trained on a UK cohort of 400,000 HF patients, leveraging self-attention mechanisms to model dynamic clinical trajectories and employing UK→US transfer learning to enhance cross-center generalizability. Key contributions include the identification—via systematic analysis—of previously underappreciated strong prognostic factors such as cancer and liver failure. TRisk significantly outperforms conventional expert-derived models (e.g., MAGGIC-EHR) in calibration, subgroup fairness, and robustness. External validation yields C-indices of 0.845 (UK cohort) and 0.802 (US cohort). SHAP-based interpretability analysis ensures clinical transparency and actionable insights.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

We developed and validated TRisk, a Transformer-based AI model predicting 36-month mortality in heart failure patients by analysing temporal patient journeys from UK electronic health records (EHR). Our study included 403,534 heart failure patients (ages 40-90) from 1,418 English general practices, with 1,063 practices for model derivation and 355 for external validation. TRisk was compared against the MAGGIC-EHR model across various patient subgroups. With median follow-up of 9 months, TRisk achieved a concordance index of 0.845 (95% confidence interval: [0.841, 0.849]), significantly outperforming MAGGIC-EHR's 0.728 (0.723, 0.733) for predicting 36-month all-cause mortality. TRisk showed more consistent performance across sex, age, and baseline characteristics, suggesting less bias. We successfully adapted TRisk to US hospital data through transfer learning, achieving a C-index of 0.802 (0.789, 0.816) with 21,767 patients. Explainability analyses revealed TRisk captured established risk factors while identifying underappreciated predictors like cancers and hepatic failure that were important across both cohorts. Notably, cancers maintained strong prognostic value even a decade after diagnosis. TRisk demonstrated well-calibrated mortality prediction across both healthcare systems. Our findings highlight the value of tracking longitudinal health profiles and revealed risk factors not included in previous expert-driven models.

Problem

• Predict 36-month mortality in heart failure patients using AI.
• Compare TRisk model performance against MAGGIC-EHR across subgroups.
• Identify underappreciated risk factors like cancers and hepatic failure.

Innovation

• Transformer-based AI model for mortality prediction
• Analyzes temporal patient journeys from EHR data
• Transfer learning adapts model to US hospital data