English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This paper presents the first deep reinforcement learning (DRL) framework toestimate the optimal Dynamic Treatment Regimes from observational medical data.This framework is more flexible and adaptive for high dimensional action andstate spaces than existing reinforcement learning methods to model real-lifecomplexity in heterogeneous disease progression and treatment choices, with thegoal of providing doctor and patients the data-driven personalized decisionrecommendations. The proposed DRL framework comprises (i) a supervised learningstep to predict the most possible expert actions, and (ii) a deep reinforcementlearning step to estimate the long-term value function of Dynamic TreatmentRegimes. Both steps depend on deep neural networks.  As a key motivational example, we have implemented the proposed framework ona data set from the Center for International Bone Marrow Transplant Research(CIBMTR) registry database, focusing on the sequence of prevention andtreatments for acute and chronic graft versus host disease aftertransplantation. In the experimental results, we have demonstrated promisingaccuracy in predicting human experts' decisions, as well as the high expectedreward function in the DRL-based dynamic treatment regimes.

Deep Reinforcement Learning for Dynamic Treatment Regimes on Medical Registry Data