Altmetric

The Artificial Intelligence Clinician learns optimal treatment strategies for sepsis in intensive care

File Description SizeFormat 
Main Text R1 290518 with figs.docxAccepted version406.04 kBMicrosoft WordView/Open
Methods R1 290518.docxSupporting information62.18 kBMicrosoft WordView/Open
Supplementary Appendix R1 290518.docxSupporting information546.72 kBMicrosoft WordView/Open
Title: The Artificial Intelligence Clinician learns optimal treatment strategies for sepsis in intensive care
Authors: Komorowski, M
Celi, LA
Badawi, O
Gordon, AC
Faisal, A
Item Type: Journal Article
Abstract: Sepsis is the third leading cause of death worldwide and the main cause of mortality in hospitals1–3, but the best treatment strategy remains uncertain. In particular, evidence suggests that current practices in the administration of intravenous fluids and vasopressors are suboptimal and likely induce harm in a proportion of patients1,4–6. To tackle this sequential decision-making problem, we developed a reinforcement learning agent, the artificial intelligence (AI) Clinician, which learns from data to predict patient dynamics given specific treatment decisions. Our agent extracted implicit knowledge from an amount of patient data that exceeds many-fold the life-time experience of human clinicians and learned optimal treatment by having analysed myriads of (mostly sub-optimal) treatment decisions. We demonstrate that the value of the AI Clinician’s selected treatment is on average reliably higher than the human clinicians. In a large validation cohort independent from the training data, mortality was lowest in patients where clinicians’ actual doses matched the AI policy. Our model provides individualized and clinically interpretable treatment decisions for sepsis that could improve patient outcomes.
Issue Date: 22-Oct-2018
Date of Acceptance: 8-Jun-2018
URI: http://hdl.handle.net/10044/1/61246
DOI: https://dx.doi.org/10.1038/s41591-018-0213-5
ISSN: 1078-8956
Publisher: Nature Publishing Group
Start Page: 1716
End Page: 1720
Journal / Book Title: Nature Medicine
Volume: 24
Copyright Statement: © 2018, Springer Nature Publishing AG
Sponsor/Funder: Engineering and Physical Sciences Research Council (EPSRC) & alumni
Keywords: Science & Technology
Life Sciences & Biomedicine
Biochemistry & Molecular Biology
Cell Biology
Medicine, Research & Experimental
Research & Experimental Medicine
INTERNATIONAL CONSENSUS DEFINITIONS
SEPTIC SHOCK
MORTALITY
MEDICINE
CRITERIA
THERAPY
ADULTS
Administration, Intravenous
Artificial Intelligence
Clinical Decision-Making
Cohort Studies
Critical Care
Female
Humans
Learning
Male
Sepsis
Software
Vasoconstrictor Agents
Humans
Sepsis
Vasoconstrictor Agents
Critical Care
Cohort Studies
Learning
Artificial Intelligence
Software
Female
Male
Administration, Intravenous
Clinical Decision-Making
11 Medical and Health Sciences
Immunology
Publication Status: Published
Appears in Collections:Faculty of Engineering
Bioengineering
Division of Surgery
Computing
Faculty of Medicine



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commons