Dynamic-DeepHit: A Deep Learning Approach for Dynamic Survival Analysis With Competing Risks Based on Longitudinal Data

2019/04/03 by Changhee Lee, Jinsung Yoon, Mihaela van der Schaar · 293 citations
Computer Science · Mathematics · Medicine · #Artificial intelligence #Computer science #Covariate #Cystic Fibrosis Research Advances #Data mining #Dynamic data #Econometrics #Event (particle physics) #Field (mathematics) #Machine Learning in Healthcare #Machine learning #Mathematics #Statistical Methods in Epidemiology #Statistical power #Statistics

paper · doi:10.1109/tbme.2019.2909027

published in IEEE Transactions on Biomedical Engineering 67(1), 122-133 (Institute of Electrical and Electronics Engineers)

openalex publication_date 2019/04/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/04

Abstract

Currently available risk prediction methods are limited in their ability to deal with complex, heterogeneous, and longitudinal data such as that available in primary care records, or in their ability to deal with multiple competing risks. This paper develops a novel deep learning approach that is able to successfully address current limitations of standard statistical approaches such as landmarking and joint modeling. Our approach, which we call Dynamic-DeepHit, flexibly incorporates the available longitudinal data comprising various repeated measurements (rather than only the last available measurements) in order to issue dynamically updated survival predictions for one or multiple competing risk(s). Dynamic-DeepHit learns the time-to-event distributions without the need to make any assumptions about the underlying stochastic models for the longitudinal and the time-to-event processes. Thus, unlike existing works in statistics, our method is able to learn data-driven associations between the longitudinal data and the various associated risks without underlying model specifications. We demonstrate the power of our approach by applying it to a real-world longitudinal dataset from the U.K. Cystic Fibrosis Registry, which includes a heterogeneous cohort of 5883 adult patients with annual follow-ups between 2009 to 2015. The results show that Dynamic-DeepHit provides a drastic improvement in discriminating individual risks of different forms of failures due to cystic fibrosis. Furthermore, our analysis utilizes post-processing statistics that provide clinical insight by measuring the influence of each covariate on risk predictions and the temporal importance of longitudinal measurements, thereby enabling us to identify covariates that are influential for different competing risks.

Cited by

• Tabular Foundation Models Can Do Survival Analysis
• Long short-term memory network with adapted attention mechanism for credit risk modeling
• Dual Model Deep Learning for Alzheimer Prognostication
• TraCeR: Transformer-Based Competing Risk Analysis with Longitudinal Covariates
• Enabling Delayed-Full Charging Through Transformer-Based Real-Time-to-Departure Modeling for EV Battery Longevity
• CLARITY: Medical World Model for Guiding Treatment Decisions by Modeling Context-Aware Disease Trajectories in Latent Space
• KANFormer for Predicting Fill Probabilities via Survival Analysis in Limit Order Books
• Frailty-Aware Transformer for Recurrent Survival Modeling of Driver Retention in Ride-Hailing Platforms
• SurvBench: A Standardised Preprocessing Pipeline for Multi-Modal Electronic Health Record Survival Analysis
• Deep Survival Analysis of Longitudinal EHR Data for Joint Prediction of Hospitalization and Death in COPD Patients
• WaveletDiff: Multilevel Wavelet Diffusion For Time Series Generation
• Think as a Doctor: An Interpretable AI Approach for ICU Mortality Prediction
• Multi-Modal Machine Learning Framework for Predicting Early Recurrence of Brain Tumors Using MRI and Clinical Biomarkers
• Dynamic Survival Prediction using Longitudinal Images based on Transformer
• Stepwise Fine and Gray: Subject-Specific Variable Selection Shows When Hemodynamic Data Improves Prognostication of Comatose Post-Cardiac Arrest Patients
• TrajSurv: Learning Continuous Latent Trajectories from Electronic Health Records for Trustworthy Survival Prediction
• SigBERT: Combining Narrative Medical Reports and Rough Path Signature Theory for Survival Risk Estimation in Oncology
• Semi-Structured Deep Piecewise Exponential Models
• Deep Survival Analysis in Multimodal Medical Data: A Parametric and Probabilistic Approach with Competing Risks
• FinSurvival: A Suite of Large Scale Survival Modeling Tasks from Finance
• A Nonlinear Low-rank Representation Model with Convolutional Neural Network for Imputing Water Quality Data
• Maximum Likelihood Estimation of Flexible Survival Densities with Importance Sampling
• Survival-oriented embeddings for improving accessibility to complex data structures
• A Causal Convolutional Low-rank Representation Model for Imputation of Water Quality Data
• Deep learning cardiac motion analysis for human survival prediction. [europepmc]
• Deep Neural Networks for Survival Analysis Using Pseudo Values. [europepmc]
• Machine Learning Algorithms to Predict Mortality and Allocate Palliative Care for Older Patients With Hip Fracture. [europepmc]
• Predicting treatment response from longitudinal images using multi-task deep learning. [europepmc]
• Intelligent Health Care: Applications of Deep Learning in Computational Medicine. [europepmc]
• Application of Machine Learning in Translational Medicine: Current Status and Future Opportunities. [europepmc]
• Long-term cancer survival prediction using multimodal deep learning. [europepmc]
• A survival model generalized to regression learning algorithms. [europepmc]
• Probabilistic Machine Learning for Healthcare. [europepmc]
• Fault Detection and Diagnosis in Industrial Processes with Variational Autoencoder: A Comprehensive Study. [europepmc]
• Prediction and Risk Stratification of Cardiovascular Disease in Diabetic Kidney Disease Patients. [europepmc]
• Developing machine learning algorithms for dynamic estimation of progression during active surveillance for prostate cancer. [europepmc]
• Prediction of survival in oropharyngeal squamous cell carcinoma using machine learning algorithms: A study based on the surveillance, epidemiology, and end results database. [europepmc]
• Dynamic Digital Twin: Diagnosis, Treatment, Prediction, and Prevention of Disease During the Life Course. [europepmc]
• Deep-learning-based prognostic modeling for incident heart failure in patients with diabetes using electronic health records: A retrospective cohort study. [europepmc]
• High-Dimensional Survival Analysis: Methods and Applications. [europepmc]
• Added prognostic value of 3D deep learning-derived features from preoperative MRI for adult-type diffuse gliomas. [europepmc]
• Dynamic prediction of malignant ventricular arrhythmias using neural networks in patients with an implantable cardioverter-defibrillator. [europepmc]
• Optimizing dynamic predictions from joint models using super learning. [europepmc]
• Incorporating longitudinal history of risk factors into atherosclerotic cardiovascular disease risk prediction using deep learning. [europepmc]
• Enhanced osteoporotic fracture prediction in postmenopausal women using Bayesian optimization of machine learning models with genetic risk score. [europepmc]
• Joint AI-driven event prediction and longitudinal modeling in newly diagnosed and relapsed multiple myeloma. [europepmc]
• Hybridizing mechanistic modeling and deep learning for personalized survival prediction after immune checkpoint inhibitor immunotherapy. [europepmc]
• Harnessing the power of longitudinal medical imaging for eye disease prognosis using Transformer-based sequence modeling. [europepmc]
• Generative AI Models in Time-Varying Biomedical Data: Scoping Review. [europepmc]

Related

• DySurv: dynamic deep learning model for survival analysis with conditional variational inference
• Dynamic Risk Prediction Triggered by Intermediate Events Using Survival Tree Ensembles
• Review for Dynamic Prediction in Clinical Survival Analysis
• Deep Survival Analysis for Competing Risk Modeling with Functional Covariates and Missing Data Imputation
• Dynamic Prediction of Competing Risk Events using Landmark Sub-distribution Hazard Model with Multiple Longitudinal Biomarker