A reinforcement learning approach for individualizing erythropoietin dosages in hemodialysis patients.
José David Martín-GuerreroFaustino J. GomezEmilio Soria-OlivasJürgen SchmidhuberMónica Climente-MartíN. Víctor Jiménez-TorresPublished in: Expert Syst. Appl. (2009)
Keyphrases
- reinforcement learning
- temporal abstractions
- clinical data
- function approximation
- patient data
- high risk
- reinforcement learning algorithms
- state space
- healthy controls
- normal controls
- learning algorithm
- medical care
- model free
- temporal difference
- markov decision processes
- multi agent
- bone marrow
- normal subjects
- optimal control
- function approximators
- dynamic programming
- medical treatment
- human resources
- optimal policy
- neurological disorders
- medical knowledge
- emergency department
- diabetes mellitus
- medical doctors
- physiological parameters