DeepSurv: Personalized Treatment Recommender System Using A Cox Proportional Hazards Deep Neural Network

Abstract

Medical practitioners use survival models to explore and understand the relationships between patients' covariates (e.g. clinical and genetic features) and the effectiveness of various treatment options. Standard survival models like the linear Cox proportional hazards model require extensive feature engineering or prior medical knowledge to model treatment interaction at an individual level. While nonlinear survival methods, such as neural networks and survival forests, can inherently model these high-level interaction terms, they have yet to be shown as effective treatment recommender systems. We introduce DeepSurv, a Cox proportional hazards deep neural network and state-of-the-art survival method for modeling interactions between a patient's covariates and treatment effectiveness in order to provide personalized treatment recommendations. We perform a number of experiments training DeepSurv on simulated and real survival data. We demonstrate that DeepSurv performs as well as or better than other state-of-the-art survival models and validate that DeepSurv successfully models increasingly complex relationships between a patient's covariates and their risk of failure. We then show how DeepSurv models the relationship between a patient's features and effectiveness of different treatment options to show how DeepSurv can be used to provide individual treatment recommendations. Finally, we train DeepSurv on real clinical studies to demonstrate how it's personalized treatment recommendations would increase the survival time of a set of patients. The predictive and modeling capabilities of DeepSurv will enable medical researchers to use deep neural networks as a tool in their exploration, understanding, and prediction of the effects of a patient's characteristics on their risk of failure.

arXiv commentPresented at the International Conference of Machine Learning Computational Biology Workshop 2016

References in corpus (4)

• Adam: A Method for Stochastic Optimization
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Random survival forests
• Self-Normalizing Neural Networks

Cited by in corpus (110)

• An overview of deep learning in medical imaging focusing on MRI
• Deep neural network models for computational histopathology: A survey
• Explainable AI for Trees: From Local Explanations to Global Understanding
• A Scalable Discrete-Time Survival Model for Neural Networks
• Deep learning cardiac motion analysis for human survival prediction
• Time-to-Event Prediction with Neural Networks and Cox Regression
• Deep Learning for Survival Analysis: A Review
• SurvSHAP(t): Time-dependent explanations of machine learning survival models
• OmiEmbed: a unified multi-task deep learning framework for multi-omics data
• Estimating heterogeneous survival treatment effect in observational data using machine learning
• DeepMTS: Deep Multi-task Learning for Survival Prediction in Patients with Advanced Nasopharyngeal Carcinoma using Pretreatment PET/CT
• A Deep Active Survival Analysis Approach for Precision Treatment Recommendations: Application of Prostate Cancer
• hist2RNA: An efficient deep learning architecture to predict gene expression from breast cancer histopathology images
• High-Dimensional Survival Analysis: Methods and Applications
• Predicting Time-to-conversion for Dementia of Alzheimer's Type using Multi-modal Deep Survival Analysis
• SurvTRACE: Transformers for Survival Analysis with Competing Events
• Variational Bayes survival analysis for unemployment modelling
• Counterfactual Phenotyping with Censored Time-to-Events
• BioFusionNet: Deep Learning-Based Survival Risk Stratification in ER+ Breast Cancer Through Multifeature and Multimodal Data Fusion
• The Concordance Index decomposition: A measure for a deeper understanding of survival prediction models
• EOCSA: Predicting Prognosis of Epithelial Ovarian Cancer with Whole Slide Histopathological Images
• CoxKAN: Kolmogorov-Arnold Networks for Interpretable, High-Performance Survival Analysis
• A Wide and Deep Neural Network for Survival Analysis from Anatomical Shape and Tabular Clinical Data
• Multi-modal Heart Failure Risk Estimation based on Short ECG and Sampled Long-Term HRV
• Individual dynamic prediction of clinical endpoint from large dimensional longitudinal biomarker history: a landmark approach
• Enabling Counterfactual Survival Analysis with Balanced Representations
• Targeted Estimation of Heterogeneous Treatment Effect in Observational Survival Analysis
• MPVNN: Mutated Pathway Visible Neural Network Architecture for Interpretable Prediction of Cancer-specific Survival Risk
• Deep Learning for Insider Threat Detection: Review, Challenges and Opportunities
• Radiomics-enhanced Deep Multi-task Learning for Outcome Prediction in Head and Neck Cancer
• DNNSurv: Deep Neural Networks for Survival Analysis Using Pseudo Values
• Federated Survival Forests
• Scaling Survival Analysis in Healthcare with Federated Survival Forests: A Comparative Study on Heart Failure and Breast Cancer Genomics
• Survival Function Matching for Calibrated Time-to-Event Predictions
• Neural Networks for Scalar Input and Functional Output
• X-CAL: Explicit Calibration for Survival Analysis
• Bounding Information Leakage in Machine Learning
• Estimating Treatment Effect under Additive Hazards Models with High-dimensional Covariates
• Large Language Model-informed ECG Dual Attention Network for Heart Failure Risk Prediction
• Real-time Mortality Prediction Using MIMIC-IV ICU Data Via Boosted Nonparametric Hazards
• BERTSurv: BERT-Based Survival Models for Predicting Outcomes of Trauma Patients
• SODEN: A Scalable Continuous-Time Survival Model through Ordinary Differential Equation Networks
• Cycling into the workshop: predictive maintenance for Barcelona's bike-sharing system
• Predicting risk of late age-related macular degeneration using deep learning
• Contrastive Learning for Predicting Cancer Prognosis Using Gene Expression Values
• CNN-based Survival Model for Pancreatic Ductal Adenocarcinoma in Medical Imaging
• Deep Survival Machines: Fully Parametric Survival Regression and Representation Learning for Censored Data with Competing Risks
• Interpretable Machine Learning for Survival Analysis
• Nearest Neighbor and Kernel Survival Analysis: Nonasymptotic Error Bounds and Strong Consistency Rates
• A General Framework for Survival Analysis and Multi-State Modelling
• LifeStream: A High-Performance Stream Processing Engine for Periodic Streams
• Disease-Atlas: Navigating Disease Trajectories with Deep Learning
• Using Geographic Location-based Public Health Features in Survival Analysis
• BigSurvSGD: Big Survival Data Analysis via Stochastic Gradient Descent
• DeepHazard: neural network for time-varying risks
• Efficient Training of Probabilistic Neural Networks for Survival Analysis
• The Brier Score under Administrative Censoring: Problems and Solutions
• Deep Kernel Survival Analysis and Subject-Specific Survival Time Prediction Intervals
• Beyond Cox Models: Assessing the Performance of Machine-Learning Methods in Non-Proportional Hazards and Non-Linear Survival Analysis
• Will Large Language Models Transform Clinical Prediction?
• An evaluation of machine learning techniques to predict the outcome of children treated for Hodgkin-Lymphoma on the AHOD0031 trial: A report from the Children's Oncology Group
• MyDigiTwin: A Privacy-Preserving Framework for Personalized Cardiovascular Risk Prediction and Scenario Exploration
• Deep Learning-Based Discrete Calibrated Survival Prediction
• The Deep Promotion Time Cure Model
• Deep Learning Approach for Predicting 30 Day Readmissions after Coronary Artery Bypass Graft Surgery
• Multivariate Arrival Times with Recurrent Neural Networks for Personalized Demand Forecasting
• Nonlinear Semi-Parametric Models for Survival Analysis
• Bayesian Weapon System Reliability Modeling with Cox-Weibull Neural Network
• ICTSurF: Implicit Continuous-Time Survival Functions with Neural Networks
• Interpretable Non-linear Survival Analysis with Evolutionary Symbolic Regression
• Probabilistic prediction of the time to hard freeze using seasonal weather forecasts and survival time methods
• Deep-CR MTLR: a Multi-Modal Approach for Cancer Survival Prediction with Competing Risks
• Dynamic prediction of time to event with survival curves
• BDNNSurv: Bayesian deep neural networks for survival analysis using pseudo values
• Neural Model-based Optimization with Right-Censored Observations
• Deep Lifetime Clustering
• AI-Based Detection, Classification and Prediction/Prognosis in Medical Imaging: Towards Radiophenomics
• Learning to rank for censored survival data
• EPIC-Survival: End-to-end Part Inferred Clustering for Survival Analysis, Featuring Prognostic Stratification Boosting
• T-SCI: A Two-Stage Conformal Inference Algorithm with Guaranteed Coverage for Cox-MLP
• Survival Mixture Density Networks
• Neural Conditional Event Time Models
• Predicting Survival Time of Ball Bearings in the Presence of Censoring
• Spectral Survival Analysis
• An RNN-Survival Model to Decide Email Send Times
• Predicting adverse outcomes following catheter ablation treatment for atrial fibrillation
• Case-Base Neural Networks: survival analysis with time-varying, higher-order interactions
• CTSL: Codebook-based Temporal-Spatial Learning for Accurate Non-Contrast Cardiac Risk Prediction Using Cine MRIs
• Uncovering life-course patterns with causal discovery and survival analysis
• Boost-R: Gradient Boosted Trees for Recurrence Data
• Metaparametric Neural Networks for Survival Analysis
• Time-driven Survival Analysis from FDG-PET/CT in Non-Small Cell Lung Cancer
• Dual Model Deep Learning for Alzheimer Prognostication
• Prediction of New Onset Diabetes after Liver Transplant
• Survival-oriented embeddings for improving accessibility to complex data structures
• DAGSurv: Directed Acyclic Graph Based Survival Analysis Using Deep Neural Networks
• Towards modelling hazard factors in unstructured data spaces using gradient-based latent interpolation
• SAFE: A Neural Survival Analysis Model for Fraud Early Detection
• DeepPrognosis: Preoperative Prediction of Pancreatic Cancer Survival and Surgical Margin via Contrast-Enhanced CT Imaging
• Simulating time to event prediction with spatiotemporal echocardiography deep learning
• Neural Diffusion Processes for Physically Interpretable Survival Prediction
• Mini-batch Estimation for Deep Cox Models: Statistical Foundations and Practical Guidance
• Ad Creative Discontinuation Prediction with Multi-Modal Multi-Task Neural Survival Networks
• Wound and episode level readmission risk or weeks to readmit: Why do patients get readmitted? How long does it take for a patient to get readmitted?
• Neural Topic Models with Survival Supervision: Jointly Predicting Time-to-Event Outcomes and Learning How Clinical Features Relate
• A meaningful prediction of functional decline in amyotrophic lateral sclerosis based on multi-event survival analysis
• Inverse-Weighted Survival Games
• Conditional Distribution Function Estimation Using Neural Networks for Censored and Uncensored Data
• Experimental Comparison of Semi-parametric, Parametric, and Machine Learning Models for Time-to-Event Analysis Through the Concordance Index
• Maximum Likelihood Estimation of Flexible Survival Densities with Importance Sampling