An experimental design for the development of adaptive treatment strategies

By Susan A. Murphy. 2005

738 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Reinforcement Learning in Healthcare: A Survey (2019, neutral): “c hypotheses and developing optimal treatments across or within groups of patients [77]. Utilizing valid data generated, for instance, from the Sequential Multiple Assignment Randomized Trial (SMART) [78], [79], an optimal DTR that is capable of optimizing the ﬁnal clinical outcome of particular int”
Deep Reinforcement Learning for Dynamic Treatment Regimes on Medical Registry Data (2017, background): “The existing DTR methods are proposed on Sequential Multiple Assignment Randomized Trial (SMART)12, in which the methods for DTR optimization are limited to clearly defined homogeneous decision stages and low-dimensional action spaces.”
Informing sequential clinical decision-making through reinforcement learning: an empirical study (2010, background): “(1) In a SMART study, the treatment actions are sampled according to a fixed (known) random exploration policy.”
Reinforcement Learning Strategies for Clinical Trials in Non-small Cell Lung Cancer (2011, background): “This design enables estimation of optimal individualized treatment regimes defined in counterfactual settings (cf. Murphy, 2005a).”
Q-Learning: Theory and Applications (2020, methodology): “That is, letting (cid:16) π n be an estimator of π opt based on a SMART of sample size n , and letting B 0 > 0, γ , α , η , (cid:19) , ζ ∈ (0,1) be constants, they derive procedures for choosing n such that (POW) There exists an α -level test of H : V ( π opt ) ≤ B based on (cid:16) π that has power”
Adaptive Interventions in Child and Adolescent Mental Health (2016, background): “Several manuscripts in this special issue describe studies that utilize a SMART design (Lavori & Dawson, 2000, 2004; Murphy, 2005).”
Precision Medicine. (2019, background): “A uthor M anuscript A uthor M anuscript A uthor M anuscript A uthor M anuscript Trials (SMARTs, Lavori and Dawson 2000, 2004; Murphy 2005a) allow for the efficient comparison of treatment sequences and fixed (i.e., not data-dependent) treatment regimes.”
New Statistical Learning Methods for Estimating Optimal Dynamic Treatment Regimes (2015, methodology): “With large number of covariates, Q- and A-learning tend to overfit the model, leading to worse performances in general.”
The International Journal of Biostatistics (2017, positive): “…as being uniquely suited to the study of these exciting questions (Thall et al. (2000), Lavori and Dawson (2000), Lavori and Dawson (2004), Murphy (2005)), with researchers in other clinical areas also beginning to implement this design (Rush et al. (2003), Schneider et al. (2006), Swartz…”
Augmented outcome-weighted learning for estimating optimal dynamic treatment regimens (2018, background): “4 Sequential multiple assignment randomized trials (SMARTs), in which randomization is implemented at each treat-ment stage, have been advocated 5 to evaluate any DTR with causal interpretation.”
Dynamic treatment regimes (2009, methodology): “Formulae for the randomization probabilities that create equal sample sizes across all dynamic treatment regimes have been provided in Murphy (2005a).”
Adaptive Interventions Treatment Modelling and Regimen Optimization Using Sequential Multiple Assignment Randomized Trials (Smart) and Q-Learning (2018, background): “For this purpose, numerous literature on clinical trial design employing experimental data are present [5, 29-32, 42, 43] that defines construction of sequential multiple assignment randomized trial (SMART) where a patient is randomized more than one time through all possible treatment options at ea”
Linear Fitted-Q Iteration with Multiple Reward Functions (2012, methodology): “Reinforcement learning methods (Szepesvári, 2010; Sutton and Barto, 1998) are already being used (Pineau et al., 2007; Murphy et al., 2007; Zhao et al., 2009) to analyze Sequential Multiple Assignment Randomized Trials (SMART) (Murphy, 2005).”
Near-Optimal Reinforcement Learning in Dynamic Treatment Regimes (2019, background): “These decision rules are alternatively known as adaptive treatment strategies [12, 13, 19, 33, 34] or treatment policies [16, 37, 38].”
Interpretable Dynamic Treatment Regimes (2016, background): “e primary role of an estimated optimal regime is to generate new scientic hypotheses or inform future research. For example, in the context of sequential multiple assignment clinical trials (SMARTs, Murphy, 2005; Lei et al., 2012) estimation of an optimal treatment regime is typically included as a ”