Reinforcement Learning for Closed-Loop Propofol Anesthesia: A Human Volunteer Study

By Brett L. Moore, Periklis Panousis, Vivekanand Kulkarni, Larry D. Pyeatt, Anthony G. Doufas. 2010

18 citations recorded by the source corpus.

Research has demonstrated the efficacy of closed-loop control of anesthesia using the bispectral index (BIS) of the electroencephalogram as the controlled variable, and the development of model-based, patient-adaptive systems has considerably improved anesthetic control. To further explore the use of model-based control in anesthesia, we investigated the application of reinforcement learning (RL) in the delivery of patient-specific, propofol-induced hypnosis in human volunteers. When compared to published performance metrics, RL control demonstrated accuracy and stability, indicating that further, more rigorous clinical study is warranted.

Subfields: Anesthesiology and Pain Medicine, Cognitive Neuroscience, Developmental Neuroscience

How researchers see this paper

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

In Anesthesiology and Pain Medicine

1. Chao Yu
2. Jiming Liu
3. Shamim Nemati
4. Brett L. Moore
5. Anthony G. Doufas
6. Larry D. Pyeatt
7. Eddy C. Borera
8. Pierre Humbert

In Cognitive Neuroscience

1. Chao Yu
2. Larry D. Pyeatt
3. Shamim Nemati
4. Jiming Liu
5. Eddy C. Borera
6. Emma Brunskill
7. Brett L. Moore
8. Anthony G. Doufas

In Developmental Neuroscience

1. Shamim Nemati
2. Chao Yu
3. Jiming Liu
4. Adam E. Gaweda

Citations in the corpus

• Reinforcement Learning in Healthcare: A Survey (2019, background): “They also presented the first clinical in vivo trial for closed-loop control of anesthesia administration using RL on 15 human volunteers [156, 157].”
• An Adaptive Neural Network Filter for Improved Patient State Estimation in Closed-Loop Anesthesia Control (2011, methodology): “technique that has been shown to deliver clinically acceptable, patient-specific control of BIS-guided hypnosis in simulation [5] and in volunteers [6].”
• On Ensuring that Intelligent Machines Are Well-Behaved (2017, methodology): “Similarly, reinforcement learning algorithms have been proposed for applications where undesirable behavior can cause financial losses [Li et al., 2010], environmental damage [Houtman et al., 2013], and even death [Moore et al., 2010].”
• Safe Reinforcement Learning (2015, background): “, 2013), or how much of a drug to give patients to optimize their prognosis (Moore et al., 2010).”
• High Confidence Policy Improvement (2015, background): “This plays a major role in precluding their use for applications in which execution of a bad policy could be costly (Jonker et al., 2004) or even dangerous (Moore et al., 2010).”