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

High-Throughput Quantitative Chemical Shift-Encoded MRI of the Liver.

Journal of magnetic resonance imaging : JMRI·2026
Same author

Comment on "Modified ultrasound-guided polidocanol sclerotherapy is an effective treatment for thyroid cystic-solid nodules: A single-center retrospective study".

American journal of otolaryngology·2026
Same author

Letter Regarding "Survival and Neurologic Performance at 30 Days and Beyond Following Out-of-Hospital Cardiac Arrest Comparing Standard Cardiopulmonary Resuscitation Versus Chest Compression-Only Resuscitation by Bystanders: A Systematic Review and Meta-Analysis".

Journal of emergency nursing·2026
Same author

Comment on: Impact of Wolbachia-infected Aedes aegypti on dengue incidence in endemic regions.

Transactions of the Royal Society of Tropical Medicine and Hygiene·2026
Same author

4D flow MRI for the evaluation of venous anatomy and flow in renal cell carcinoma with venous tumor thrombus: a pictorial essay.

Abdominal radiology (New York)·2026
Same author

Methodological considerations in estimating the burden of congenital heart disease among Indian children.

Indian journal of thoracic and cardiovascular surgery·2026

Related Experiment Video

Updated: Oct 3, 2025

In vitro Assessment of Aortic Regurgitation Using Four-Dimensional Flow Magnetic Resonance Imaging
11:16

In vitro Assessment of Aortic Regurgitation Using Four-Dimensional Flow Magnetic Resonance Imaging

Published on: February 25, 2022

3.5K

Fully automated intracardiac 4D flow MRI post-processing using deep learning for biventricular segmentation.

Philip A Corrado1, Andrew L Wentland2, Jitka Starekova2

  • 1University of Wisconsin-Madison, 1111 Highland Ave, Madison, WI, 53705, USA. pcorrado2@wisc.edu.

European Radiology
|February 17, 2022
PubMed
Summary

Automated deep learning segmentation of cardiac MRI significantly reduces post-processing time and variability in hemodynamic measurements. This advanced technique matches manual segmentation accuracy, improving efficiency for cardiovascular assessments.

Keywords:
Deep learningHeart ventriclesHemodynamicsMagnetic resonance imagingObserver variation

More Related Videos

Human Fetal Blood Flow Quantification with Magnetic Resonance Imaging and Motion Compensation
06:56

Human Fetal Blood Flow Quantification with Magnetic Resonance Imaging and Motion Compensation

Published on: January 7, 2021

2.5K
Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging
10:44

Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging

Published on: June 21, 2024

664

Related Experiment Videos

Last Updated: Oct 3, 2025

In vitro Assessment of Aortic Regurgitation Using Four-Dimensional Flow Magnetic Resonance Imaging
11:16

In vitro Assessment of Aortic Regurgitation Using Four-Dimensional Flow Magnetic Resonance Imaging

Published on: February 25, 2022

3.5K
Human Fetal Blood Flow Quantification with Magnetic Resonance Imaging and Motion Compensation
06:56

Human Fetal Blood Flow Quantification with Magnetic Resonance Imaging and Motion Compensation

Published on: January 7, 2021

2.5K
Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging
10:44

Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging

Published on: June 21, 2024

664

Area of Science:

  • Cardiovascular Imaging
  • Medical Artificial Intelligence
  • Biomedical Engineering

Background:

  • 4D flow MRI offers comprehensive intracardiac blood flow assessment for cardiovascular diseases.
  • Current post-processing relies on time-consuming and subjective manual ventricular segmentation.
  • This limits the clinical utility of 4D flow MRI due to processing demands and interobserver variability.

Purpose of the Study:

  • To evaluate a deep learning (DL) network for automated left and right ventricular segmentation in 4D flow MRI.
  • To compare the accuracy and variability of DL-based segmentation with manual segmentation for hemodynamic parameters.
  • To assess the potential of DL to streamline 4D flow MRI post-processing.

Main Methods:

  • A DL network was fine-tuned and applied to segment ventricles in 26 subjects.
  • Automated segmentation was compared against manual segmentation by three observers using Dice scores and relative deviations.
  • Segmentation was mapped to 4D flow data to assess hemodynamic parameters like kinetic energy.

Main Results:

  • Automated segmentation achieved high Dice scores (LV: 0.92, RV: 0.86), comparable to interobserver variability (LV: 0.91, RV: 0.87).
  • Relative deviations in kinetic energy measurements were lower for automated vs. manual (LV: 8%, RV: 15%) compared to interobserver deviations (LV: 11%, RV: 19%).
  • The DL method demonstrated comparable or superior agreement to manual segmentation for hemodynamic analysis.

Conclusions:

  • Fully automated post-processing of intraventricular 4D flow MRI is feasible using deep learning.
  • Hemodynamic measurements derived from automated segmentation show variability similar to or less than interobserver variability.
  • This automated approach significantly accelerates 4D flow MRI post-processing and enhances repeatability.