Key result
A random survival forest model predicted 1-year mortality in patients with cardiac arrhythmias with a c-statistic of 0.81, outperforming the traditional Cox proportional hazard model (c-statistic 0.733).
Why the study?
Does a random survival forest model improve prediction accuracy of 1-year mortality compared to a Cox proportional hazard model in patients with cardiac arrhythmias?
Population
10,488 cardiac arrhythmias patients from the public MIMIC II clinical database
Comparison
Random survival forest model using 40 risk… vs Traditional Cox proportional hazard (CPH) model
Design
Cohort
Follow-up
1-year
Authors
Loading...
May enhance arrhythmia mortality prediction; hypothesis-generating and requires prospective validation before clinical use.
Observational (n=10,488)
Does a random survival forest model improve prediction accuracy of 1-year mortality compared to a Cox proportional hazard model in patients with cardiac arrhythmias?
Absolute Event Rate: 0.81% vs 0.733%
A random survival forest model significantly outperforms traditional Cox proportional hazard models in predicting 1-year mortality among patients with cardiac arrhythmias by accounting for nonlinear impacts of risk factors.
Miao et al. (2015) conducted an observational in Cardiac arrhythmias (n=10,488). Random survival forest (RSF) model vs. Cox proportional hazard approach (CPH) was evaluated on Prediction accuracy of 1-year all-cause mortality measured by c-statistic. A random survival forest model predicted 1-year mortality in patients with cardiac arrhythmias with a c-statistic of 0.81, outperforming the traditional Cox proportional hazard model (c-statistic 0.733).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: