Random Survival Forest Versus Elastic-Net Regularized Cox Regression for Survival Prediction in Acute Myeloid Leukemia at Distinct Treatment Time Points: Model Performance Comparison Study

随机生存森林与弹性网络正则化Cox回归在不同治疗时间点对急性髓系白血病生存预测的模型性能比较研究

阅读:1

Abstract

BACKGROUND: Risk group stratification based on the prediction of survival of patients with acute myeloid leukemia (AML) is complex. Despite common risk group categorization guidelines, the overall prognosis remains poor. Machine learning techniques have been shown to provide more accurate risk group stratification than conventional approaches using trial data. However, many time-to-event (TTE) models do not use training sets constrained to specific time windows, instead using aggregations of trial data. OBJECTIVE: This study aimed to evaluate the performance of (1) random survival forest (RSF) and (2) Cox proportional hazard regression with elastic net regularization (CoxNet) for survival prediction of patients with AML within a censoring window trained with available data recorded at discrete time points during the United Kingdom National Cancer Research Institute Acute Myeloid Leukaemia 17 randomized controlled trial (AML17). METHODS: For each stage in the AML17 trial, separate models were trained for each exhaustive k-choice combination of available AML17 data subsets. Data combinations for each model were further constrained according to the respective trial stage to avoid data leakage. Preliminary Pearson correlation methods were used to remove directly correlating features with the TTE prediction (time-to-death/5-y censoring point). Repeated k-fold stratified cross-validation was used on each dataset ablation to find candidate models. Permutation importance and elastic net regularization were used to monitor stability across validation folds and reduce the feature set of the highest performing stage RSF and Cox proportional hazard regression models, respectively. Finally, selected ablated models were re-evaluated using the nested, k-fold, stratified sampling cross-validation method with bootstrapping. RESULTS: Concordance index ranked the best models for data constricted up to the end of induction (RSF=0.68, CoxNet=0.67), stages 1 (RSF=0.69, CoxNet=0.68), 2 (RSF=0.68, CoxNet=0.66), and 3 (RSF=0.69, CoxNet=0.63) of the trial. CONCLUSIONS: This study details the high prediction accuracy for time-to-survival-event predictions when training sets of CoxNet and RSF models, which are sequentially constricted to data measured up to the end of respective AML17 trial stages. The performance of these sequential TTE models is intended to justify their use as part of a wider digital twin system simulating multiple TTE outcomes for patients with AML.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。