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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

From Clinical Evidence to Primary Care Practice: Group Exercise and Nutrition Interventions to Support Bone Density, Fall Prevention, and Fracture Risk Reduction.

Applied physiology, nutrition, and metabolism = Physiologie appliquee, nutrition et metabolisme·2026
Same author

An ancestry-enriched HNF4A variant and GP2 reveal distinct mechanisms of type 2 diabetes in exome-wide study of 13,674 cases and 41,024 controls.

medRxiv : the preprint server for health sciences·2026
Same author

Exploring the perioperative experiences of youth undergoing cardiac surgery: A qualitative needs assessment for tailoring a mobile self-management app.

Journal of clinical and translational science·2026
Same author

Does vesicoureteral reflux severity impact breakthrough UTI risk: a multi-institutional analysis and risk stratification.

World journal of urology·2026
Same author

Closed-loop stimulation modulates attention shifting in children.

Nature neuroscience·2026
Same author

Posttraumatic Symptoms as Predictors of Engagement With a Mobile App for Coping After Military Sexual Trauma: Public Usage Data Analysis Study.

Journal of medical Internet research·2026

Related Experiment Video

Updated: Sep 21, 2025

Evaluation of Right Ventricular Function in Experimental Models of Pulmonary Arterial Hypertension
10:03

Evaluation of Right Ventricular Function in Experimental Models of Pulmonary Arterial Hypertension

Published on: June 27, 2025

342

Retraining an Artificial Intelligence Algorithm to Calculate Left Ventricular Ejection Fraction in Pediatrics.

Mael Zuercher1, Steven Ufkes2, Lauren Erdman2

  • 1Department of Anesthesia and Pain Medicine, The Hospital for Sick Children, Toronto, Ontario, Canada; Department of Anesthesia, University of Toronto, Toronto, Ontario, Canada.

Journal of Cardiothoracic and Vascular Anesthesia
|May 31, 2022
PubMed
Summary

A deep-learning algorithm was fine-tuned to accurately calculate left ventricular ejection fraction (LVEF) in pediatric patients. The refined AI model achieved clinically acceptable error rates, aiding in LVEF assessment for children.

Keywords:
artificial intelligencecardiac functiondeep learningpediatric patients

More Related Videos

Lumped-Parameter and Finite Element Modeling of Heart Failure with Preserved Ejection Fraction
09:20

Lumped-Parameter and Finite Element Modeling of Heart Failure with Preserved Ejection Fraction

Published on: February 13, 2021

6.6K
Echocardiographic Assessment of the Right Heart in Mice
09:29

Echocardiographic Assessment of the Right Heart in Mice

Published on: November 27, 2013

23.2K

Related Experiment Videos

Last Updated: Sep 21, 2025

Evaluation of Right Ventricular Function in Experimental Models of Pulmonary Arterial Hypertension
10:03

Evaluation of Right Ventricular Function in Experimental Models of Pulmonary Arterial Hypertension

Published on: June 27, 2025

342
Lumped-Parameter and Finite Element Modeling of Heart Failure with Preserved Ejection Fraction
09:20

Lumped-Parameter and Finite Element Modeling of Heart Failure with Preserved Ejection Fraction

Published on: February 13, 2021

6.6K
Echocardiographic Assessment of the Right Heart in Mice
09:29

Echocardiographic Assessment of the Right Heart in Mice

Published on: November 27, 2013

23.2K

Area of Science:

  • Cardiology
  • Artificial Intelligence in Medicine
  • Pediatric Echocardiography

Background:

  • Accurate measurement of left ventricular ejection fraction (LVEF) is crucial for diagnosing and managing pediatric cardiac conditions.
  • Current methods for LVEF calculation can be subject to inter-observer variability and require specialized expertise.
  • Developing automated tools for LVEF assessment can improve consistency and accessibility in pediatric cardiology.

Purpose of the Study:

  • To adapt and validate a deep-learning algorithm (EchoNet-Dynamic) for calculating LVEF in pediatric patients.
  • To achieve a mean absolute error (MAE) of ≤5% for LVEF calculation in the pediatric population.
  • To assess the clinical feasibility of using artificial intelligence for pediatric LVEF monitoring.

Main Methods:

  • Retraining and fine-tuning of the adult-based EchoNet-Dynamic algorithm using a diverse dataset of pediatric echocardiograms.
  • Validation against LVEF values determined by expert clinical interpretation (gold standard).
  • Statistical analysis including Bland-Altman plots to evaluate model accuracy and agreement.

Main Results:

  • The fine-tuned algorithm demonstrated a significantly improved MAE of 4.47% with R²=0.87, meeting the predefined clinical target.
  • The model showed a slight underestimation of LVEF (bias = -2.42%) with 95% limits of agreement from -12.32% to 7.47%.
  • The AI model successfully calculated LVEF within clinically acceptable error margins across a wide age range of pediatric patients.

Conclusions:

  • The fine-tuned deep-learning model provides a reliable method for calculating pediatric LVEF within clinically acceptable error.
  • This AI tool has the potential to reduce operator-dependent errors in LVEF measurements.
  • The algorithm can support LVEF assessment by less experienced users, enhancing diagnostic capabilities in pediatric care.