Finding the Optimal Dynamic Treatment Regime Using Smooth Fisher Consistent Surrogate Loss

DOI10.1214/24-AOS2363MaRDI QIDQ6382200zbMATH OpenOpenAlexFDO

Authors Nilanjana Laha, Aaron Sonabend-W, Rajarshi Mukherjee, Tianxi Cai

Publication date 3 November 2021

Full work available at URL https://doi.org/10.1214/24-aos2363

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Applications of statistics to biology and medical sciences; meta analysis (62P10) Learning and adaptive systems in artificial intelligence (68T05)

Abstract: Large health care data repositories such as electronic health records (EHR) opens new opportunities to derive individualized treatment strategies to improve disease outcomes. We study the problem of estimating sequential treatment rules tailored to patient's individual characteristics, often referred to as dynamic treatment regimes (DTRs). We seek to find the optimal DTR which maximizes the discontinuous value function through direct maximization of a fisher consistent surrogate loss function. We show that a large class of concave surrogates fails to be Fisher consistent, which differs from the classic setting for binary classification. We further characterize a non-concave family of Fisher consistent smooth surrogate functions, which can be optimized with gradient descent using off-the-shelf machine learning algorithms. Compared to the existing direct search approach under the support vector machine framework (Zhao et al., 2015), our proposed DTR estimation via surrogate loss optimization (DTRESLO) method is more computationally scalable to large sample size and allows for a broader functional class for the predictor effects. We establish theoretical properties for our proposed DTR estimator and obtain a sharp upper bound on the regret corresponding to our DTRESLO method. Finite sample performance of our proposed estimator is evaluated through extensive simulations and an application on deriving an optimal DTR for treatment sepsis using EHR data from patients admitted to intensive care units.

Recommendations

• New statistical learning methods for estimating optimal dynamic treatment regimes
• Multicategory Angle-Based Learning for Estimating Optimal Dynamic Treatment Regimes With Censored Data
• Optimal Dynamic Treatment Regimes
• C-learning: a new classification framework to estimate optimal dynamic treatment regimes
• Generalization error bounds of dynamic treatment regimes in penalized regression-based learning

Cited in

(3)

• Estimation of optimal dynamic treatment assignment rules under policy constraints
• Federated learning of robust individualized decision rules with application to heterogeneous multihospital sepsis population
• Stage-aware learning for dynamic treatments

This page was built for publication: Finding the Optimal Dynamic Treatment Regime Using Smooth Fisher Consistent Surrogate Loss

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6382200)