Optimal Tuning of Random Survival Forest Hyperparameter with an Application to Liver Disease

Abstract

Background: Random Forest (RF) is a technique that optimises predictive accuracy by fitting an ensemble of trees to stabilise model estimates. The RF techniques were adapted into survival analysis to model the survival of patients with liver disease in order to identify biomarkers that are highly influential in patient prognostics. Methods: The methodology of this study begins by applying the classical Cox proportional hazard (Cox-PH) model and three parametric survival models (exponential, Weibull and lognormal) to the published dataset. The study further applied the supervised learning methods of Tuning Random Survival Forest (TRSF) parameters and the conditional inference Forest (Cforest) to optimally predict patient survival probabilities. Results: The efficiency of these models was compared using the Akaike information criteria (AIC) and integrated Brier score (IBS). The results revealed that the Cox-PH model (AIC = 185.7233) outperforms the three classical models. We further analysed these data to observe the functional relationships that exist between the patient survival function and the covariates using TRSF. Conclusion: The IBS result of the TRFS demonstrated satisfactory performance over other methods. Ultimately, it was observed from the TRSF results that some of the covariates contributed positively and negatively to patient survival prognostics.

A New Generalized Gamma-Weibull Distribution with Applications to Time-to-event Data

“…Taking a broader perspective, [6] delved into a generalized form of the log-transformed inverse Weibull distribution, conducting an extensive investigation of its theoretical properties. In summary, several studies in the literature, such as those cited in references ( [7,8]), have explored and developed flexible distribution models capable of handling highly skewed time-to-event data. However, there still remains a need for the development of a distribution that can adeptly accommodate highly skewed time-to-event data while simultaneously considering censoring information.…”