Idioma:

Outcome Weighted Learning

Pan, Yinghao ; Zhao, Ying‐Qi Balakrishnan, N. ; Everitt, Brian ; Colton, Theodore ; Teugels, Jozef L. ; Piegorsch, Walter ; Ruggeri, Fabrizio

Wiley StatsRef: Statistics Reference Online, 2019, p.1-9

Texto completo disponível

Citações Citado por

Título:
Outcome Weighted Learning
Autor: Pan, Yinghao ; Zhao, Ying‐Qi
Balakrishnan, N. ; Everitt, Brian ; Colton, Theodore ; Teugels, Jozef L. ; Piegorsch, Walter ; Ruggeri, Fabrizio
Assuntos: Applications ; Statistics in Biological and Medical Sciences
É parte de: Wiley StatsRef: Statistics Reference Online, 2019, p.1-9
Descrição: Due to patient's heterogeneous response to treatment, there is a growing demand to develop treatment strategies according to individual characteristics, which could lead to better health outcomes. Individualized treatment rules aim to identify if, when, which, and to whom treatment should be applied. In this article, we review a general framework of outcome weighted learning, a recently proposed machine learning technique for estimating optimal individualized treatment rules. The methods developed within this framework can be applied to both a single‐stage setup, where a single treatment decision is to be made, and multi‐stage setups, where a sequence of treatment decisions will be made. Furthermore, the methods can handle different types of outcomes, including continuous, categorical, and survival outcomes, and various treatment formations such as binary/ordinal treatments and continuous dose level. Finally, we briefly discuss some recent developments in this research area.
Editor: Chichester, UK: John Wiley & Sons, Ltd
Idioma: Inglês

Voltar para lista de resultados