Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Rule-base derivation for intensive care ventilator control using ANFIS.

H F Kwok1, D A Linkens, M Mahfouf

  • 1Department of Automatic Control and Systems Engineering, University of Sheffield, Mappin Street, Sheffield S1 3JD, UK.

Artificial Intelligence in Medicine
|December 6, 2003
PubMed
Summary

Adaptive neuro-fuzzy inference system (ANFIS) effectively derives ventilator control rules, reducing expert time. ANFIS models clinician decisions better than previous fuzzy advisors, showing comparable performance in simulations.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Effect of Intraoperative High Positive End-Expiratory Pressure (PEEP) With Recruitment Maneuvers vs Low PEEP on Postoperative Pulmonary Complications in Obese Patients: A Randomized Clinical Trial.

JAMA·2019
Same author

Correction to: Protective ventilation with high versus low positive end-expiratory pressure during one-lung ventilation for thoracic surgery (PROTHOR): study protocol for a randomized controlled trial.

Trials·2019
Same author

Protective ventilation with high versus low positive end-expiratory pressure during one-lung ventilation for thoracic surgery (PROTHOR): study protocol for a randomized controlled trial.

Trials·2019
Same author

Association between night-time surgery and occurrence of intraoperative adverse events and postoperative pulmonary complications.

British journal of anaesthesia·2019
Same author

Intraoperative ventilation settings and their associations with postoperative pulmonary complications in obese patients.

British journal of anaesthesia·2018
Same author

Problem behaviours and psychotropic medication use in intellectual disability: a multinational cross-sectional survey.

Journal of intellectual disability research : JIDR·2018

Area of Science:

  • Medical Artificial Intelligence
  • Computational Medicine
  • Intelligent Control Systems

Background:

  • Fuzzy systems are increasingly used in medicine, but rule-base derivation is time-consuming.
  • Traditional methods require extensive expert input, posing a bottleneck in development.

Purpose of the Study:

  • To present the adaptive neuro-fuzzy inference system (ANFIS) for efficient rule-base derivation in ventilator control.
  • To model clinician decision-making in adjusting inspired fraction of oxygen (FiO2).

Main Methods:

  • ANFIS and Multilayer Perceptron (MLP) modeled relationships between arterial oxygen tension (PaO2), FiO2, and positive end-expiratory pressure (PEEP).
  • Clinician advice from 71 scenarios was used to train and compare models against a previous fuzzy advisor (FAVeM) and a radial basis network (RBN-MB).

Related Experiment Videos

Main Results:

  • ANFIS (0.694) and MLP (0.701) showed better correlation with clinician decisions than FAVeM (0.630).
  • In closed-loop simulations, ANFIS (0.852), MLP (0.962), and RBN-MB (0.787) performed comparably to clinicians, while FAVeM (0.332) differed but maintained safety limits.

Conclusions:

  • ANFIS offers an efficient alternative for deriving ventilator control rules, reducing reliance on extensive expert input.
  • ANFIS and MLP demonstrate strong potential for modeling clinical expertise in mechanical ventilation.