LeBlanc Michael, Moon James, Crowley John
Fred Hutchinson Cancer Research Center, 1100 Fairview Avenue North, M3-C102, Seattle, Washington 98109, USA.
Biometrics. 2005 Jun;61(2):370-8. doi: 10.1111/j.1541-0420.2005.020738.x.
We construct interpretable prognostic rules based on a sequence of "box-shaped" regions in the predictor space indexed by the fraction of patients in the prognostic group. In addition, the method can be used as a building block to construct more general prognostic rules based on unions of boxes, or even as a tool to find multiple prognostic groups. Simulations are used to study the properties of the new method and compare it to constructing prognostic groups based on regression trees and linear proportional hazards (PH) models. We consider an example based on data from several completed clinical trials for patients with multiple myeloma.
我们基于预测空间中一系列由预后组患者比例索引的“盒状”区域构建可解释的预后规则。此外，该方法可用作构建基于盒子并集的更通用预后规则的基础模块，甚至可作为寻找多个预后组的工具。通过模拟研究新方法的性质，并将其与基于回归树和线性比例风险（PH）模型构建预后组的方法进行比较。我们考虑一个基于多个已完成的多发性骨髓瘤患者临床试验数据的例子。
