Non-parametric recurrent events analysis with BART and an application to the hospital admissions of patients with diabetes

Sparapani, Rodney; Rein, Lisa; Tarima, Sergey; Jackson, Tourette A.; Meurer, John R.

doi:10.1093/biostatistics/kxy032

Cited by 10 publications

(8 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This black‐box relationship may not be readily interpretable which is a barrier to adoption in clinical practice. A strategy to achieve an interpretable rule is the fit‐the‐fit method which is an adaption of decoupling, shrinkage, and selection 32 to BART 33 (employed in the HIV AIDS example; see Section 6.2). For fit‐the‐fit with precision medicine, a single tree ( H = 1) is fit to the counter‐factual differences computed from the ensemble model, that is, the data for the single tree is based on the predictions from the black‐box model.…”

Section: Precision Medicinementioning

confidence: 99%

Nonparametric machine learning for precision medicine with longitudinal clinical trials and Bayesian additive regression trees with mixed models

Spanbauer

Sparapani

2021

Statistics in Medicine

View full text Add to dashboard Cite

Precision medicine is an active area of research that could offer an analytic paradigm shift for clinical trials and the subsequent treatment decisions based on them. Clinical trials are typically analyzed with the intent of discovering beneficial treatments if the same treatment is applied to the entire population under study. But, such a treatment strategy could be suboptimal if subsets of the population exhibit varying treatment effects. Identifying subsets of the population experiencing differential treatment effect and forming individualized treatment rules is a task well‐suited to modern machine learning methods such as tree‐based ensemble predictive models. Specifically, Bayesian additive regression trees (BART) has shown promise in this regard because of its exceptional performance in out‐of‐sample prediction. Due to the unique inferential needs of precision medicine for clinical trials, we have proposed the BART extensions explicated here. We incorporate random effects for longitudinal repeated measures and subject clustering within medical centers. The addition of a novel interaction detection prior to identify treatment heterogeneity among clinical trial patients and its association with patient characteristics. These extensions are unified under a framework that we call mixedBART. Simulation studies and applications of precision medicine based on real randomized clinical trials data examples are presented.

show abstract

Section: Precision Medicinementioning

confidence: 99%

Nonparametric machine learning for precision medicine with longitudinal clinical trials and Bayesian additive regression trees with mixed models

Spanbauer

Sparapani

2021

Statistics in Medicine

View full text Add to dashboard Cite

show abstract

“…The logit normal prior is incorporated into the overall BART algorithm, resulting in a nonlinear prediction method that is grounded entirely in Bayesian probability. Furthermore, this prior can be incorporated into BART for any type of outcome including probit, logistic, survival (Sparapani et al, 2016), competing risks (Sparapani et al, 2020), recurrent events (Sparapani et al, 2020), and repeated measures/random effects (Spanbauer & Sparapani, 2021; Tan et al, 2018) resulting in broad applicability of this method.…”

Section: Introductionmentioning

confidence: 99%

Sparse prediction informed by genetic annotations using the logit normal prior for Bayesian regression tree ensembles

Spanbauer

Pan

2022

Genetic Epidemiology

View full text Add to dashboard Cite

Using high‐dimensional genetic variants such as single nucleotide polymorphisms (SNP) to predict complex diseases and traits has important applications in basic research and other clinical settings. For example, predicting gene expression is a necessary first step to identify (putative) causal genes in transcriptome‐wide association studies. Due to weak signals, high‐dimensionality, and linkage disequilibrium (correlation) among SNPs, building such a prediction model is challenging. However, functional annotations at the SNP level (e.g., as epigenomic data across multiple cell‐ or tissue‐types) are available and could be used to inform predictor importance and aid in outcome prediction. Existing approaches to incorporate annotations have been based mainly on (generalized) linear models. Bayesian additive regression trees (BART), in contrast, is a reliable method to obtain high‐quality nonlinear out of sample predictions without overfitting. Unfortunately, the default prior from BART may be too inflexible to handle sparse situations where the number of predictors approaches or surpasses the number of observations. Motivated by our real data application, this article proposes an alternative prior based on the logit normal distribution because it provides a framework that is adaptive to sparsity and can model informative functional annotations. It also provides a framework to incorporate prior information about the between SNP correlations. Computational details for carrying out inference are presented along with the results from a simulation study and a genome‐wide prediction analysis of the Alzheimer's Disease Neuroimaging Initiative data.

show abstract

“…To resolve the aforementioned issues, in this article we introduce BART‐PS, a flexible Bayesian continuous monitoring design that handles the noncompliance issue under the Principal Stratification framework using Bayesian additive regression trees (BARTs) 26 . BART has been widely used in many application including survival analysis, 27 molecular biology, 28 and disease studies 29 thanks to its excellent performance in prediction and variable selection 30 . Within BART‐PS, we replace the one‐sided access assumption by the more realistic monotonicity assumption, and use BART for the simultaneous identification of useful compliance‐related covariates and the prediction of patients' compliance stratum.…”

Section: Introductionmentioning

confidence: 99%

Bayesian phase II clinical trial design with noncompliance

Ren

Shen

Zhang

et al. 2021

Statistics in Medicine

View full text Add to dashboard Cite

Noncompliance issue is common in early phase clinical trials; and may lead to biased estimation of the intent‐to‐treat effect and incorrect conclusions for the clinical trial. In this work, we propose a Bayesian approach for sequentially monitoring the phase II randomized clinical trials that takes account for the noncompliance information. We adopt the principal stratification framework and propose to use Bayesian additive regression trees for selecting useful baseline covariates and estimating the complier average causal effect (CACE) for both efficacy and toxicity outcomes. The decision of early termination or not is then made adaptively based on the estimated CACE from the accumulated data. Simulation studies have confirmed the excellent performance of the proposed design in the presence of noncompliance.

show abstract

Non-parametric recurrent events analysis with BART and an application to the hospital admissions of patients with diabetes

Cited by 10 publications

References 33 publications

Nonparametric machine learning for precision medicine with longitudinal clinical trials and Bayesian additive regression trees with mixed models

Nonparametric machine learning for precision medicine with longitudinal clinical trials and Bayesian additive regression trees with mixed models

Sparse prediction informed by genetic annotations using the logit normal prior for Bayesian regression tree ensembles

Bayesian phase II clinical trial design with noncompliance

Contact Info

Product

Resources

About