Abstract
BackgroundClinical research and medical practice can be advanced through the prediction of an individual’s health state, trajectory, and responses to treatments. However, the majority of current clinical risk prediction models are based on regression approaches or machine learning algorithms that are static, rather than dynamic. To benefit from the increasing emergence of large, heterogeneous data sets, such as electronic health records (EHRs), novel tools to support improved clinical decision making through methods for individual-level risk prediction that can handle multiple variables, their interactions, and time-varying values are necessary.MethodsWe introduce a novel dynamic approach to clinical risk prediction for survival, longitudinal, and multivariate (SLAM) outcomes, called random forest for SLAM data analysis (RF-SLAM). RF-SLAM is a continuous-time, random forest method for survival analysis that combines the strengths of existing statistical and machine learning methods to produce individualized Bayes estimates of piecewise-constant hazard rates. We also present a method-agnostic approach for time-varying evaluation of model performance.ResultsWe derive and illustrate the method by predicting sudden cardiac arrest (SCA) in the Left Ventricular Structural (LV) Predictors of Sudden Cardiac Death (SCD) Registry. We demonstrate superior performance relative to standard random forest methods for survival data. We illustrate the importance of the number of preceding heart failure hospitalizations as a time-dependent predictor in SCA risk assessment.ConclusionsRF-SLAM is a novel statistical and machine learning method that improves risk prediction by incorporating time-varying information and accommodating a large number of predictors, their interactions, and missing values. RF-SLAM is designed to easily extend to simultaneous predictions of multiple, possibly competing, events and/or repeated measurements of discrete or continuous variables over time.Trial registration: LV Structural Predictors of SCD Registry (clinicaltrials.gov, NCT01076660), retrospectively registered 25 February 2010
Highlights
Clinical research and medical practice can be advanced through the prediction of an individual’s health state, trajectory, and responses to treatments
Methods: random forest for survival, longitudinal, and multivariate (RF-SLAM) data analysis To overcome the limitations of current sudden cardiac arrest (SCA) risk modeling approaches, we develop Random Forest for Survival, Longitudinal, and Multivariate (RF-SLAM) data analysis, a method that builds upon the concept of decision trees for risk stratification
We introduce an extension of the random forest methodology, which we call random forest for SLAM data analysis (RF-SLAM), based upon the Poisson regression log-likelihood as the split statistic to allow for the inclusion of time-varying predictors and the analysis of survival data without the restrictive proportional hazards assumption
Summary
Clinical research and medical practice can be advanced through the prediction of an individual’s health state, trajectory, and responses to treatments. Clinical risk assessment has been a long-standing challenge in medicine, at the individual level [1] Questions such as “what is the probability that this patient has a particular disease?” or “what is the probability that this patient will benefit from a particular treatment?” are difficult to answer objectively but are essential in order to Recent advances in biomedical, information, and communication technologies increase the potential to substantially improve clinical risk prediction. Even when variables are repeatedly measured on the same individual over time, it is common to base the patient’s risk score only on the last available measurement rather than the full history of measurements This practice is inconsistent with the inherently dynamic nature of human health and disease; it discards valuable information from the history of the predictors, such as the rate of change of the variables or the occurrence of prior events [1]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
