Optimization of anemia treatment in hemodialysis patients via reinforcement learning
作者：
Highlights：
• We use batch reinforcement learning for optimizing the treatment of renal anemia.
• The proposed method learns optimal treatments from medical records.
• A standard dose adjustment protocol is compared with the learned policy.
• Treatments are tailored according to the patients’ characteristics.
摘要
Highlights•We use batch reinforcement learning for optimizing the treatment of renal anemia.•The proposed method learns optimal treatments from medical records.•A standard dose adjustment protocol is compared with the learned policy.•Treatments are tailored according to the patients’ characteristics.
论文关键词：Reinforcement learning,Markov decision processes,Fitted Q iteration,Chronic kidney disease,Renal anemia,Darbepoietin alfa
论文评审过程：Received 3 November 2013, Revised 23 June 2014, Accepted 11 July 2014, Available online 19 July 2014.
论文官网地址：https://doi.org/10.1016/j.artmed.2014.07.004
