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

Identification of ischemic stroke risk in patients with left ventricular non-compaction using echocardiography, deep learning, and radiomics.

BMC medical imaging·2026
Same author

Bioinformatics and network pharmacology to explore Huangqi Guizhi Wuwu Decoction in regulating mitophagy to ameliorate doxorubicin-induced cardiotoxicity the potential mechanism.

Medicine·2026
Same author

Artificial Intelligence-Assisted Wireless Handheld Ultrasound for Screening Developmental Dysplasia of the Hip in Infants.

Journal of ultrasound in medicine : official journal of the American Institute of Ultrasound in Medicine·2026
Same author

Sediment-porewater dynamics of neurotoxin domoic acid in marginal seas of northern China: Seasonal prevalence and ecological risks in mariculture bays.

Harmful algae·2026
Same author

Formononetin and Rhein from Bitong Mixture Alleviate Rheumatoid Arthritis-Related Inflammation: An Integrated WGCNA, Machine Learning and In Vitro Study.

Pharmaceuticals (Basel, Switzerland)·2026
Same author

The spatial distribution, source analysis and ecological risk assessment of organochlorine pesticides (OCPs) in the Bohai Bay, China.

Marine pollution bulletin·2026

Related Experiment Video

Updated: Jun 14, 2025

Echocardiographic Approaches and Protocols for Comprehensive Phenotypic Characterization of Valvular Heart Disease in Mice
12:12

Echocardiographic Approaches and Protocols for Comprehensive Phenotypic Characterization of Valvular Heart Disease in Mice

Published on: February 14, 2017

16.0K

Automated echocardiographic diastolic function grading: A hybrid multi-task deep learning and machine learning

Qizhe Cai1, Mingming Lin1, Miao Zhang1

  • 1Department of Ultrasound, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China.

International Journal of Cardiology
|September 1, 2024
PubMed
Summary
This summary is machine-generated.

A new hybrid deep learning and machine learning algorithm, MMnet, automates left ventricular diastolic function (LVDF) assessment using echocardiography. This efficient tool accurately grades diastolic function, improving upon traditional methods.

Keywords:
Deep learningLeft ventricular diastolic functionMachine learningMulti-task out-put

More Related Videos

Ultrasonic Assessment of Myocardial Microstructure
10:53

Ultrasonic Assessment of Myocardial Microstructure

Published on: January 14, 2014

5.4K
Evaluation of Left Ventricular Structure and Function using 3D Echocardiography
06:34

Evaluation of Left Ventricular Structure and Function using 3D Echocardiography

Published on: October 28, 2020

3.9K

Related Experiment Videos

Last Updated: Jun 14, 2025

Echocardiographic Approaches and Protocols for Comprehensive Phenotypic Characterization of Valvular Heart Disease in Mice
12:12

Echocardiographic Approaches and Protocols for Comprehensive Phenotypic Characterization of Valvular Heart Disease in Mice

Published on: February 14, 2017

16.0K
Ultrasonic Assessment of Myocardial Microstructure
10:53

Ultrasonic Assessment of Myocardial Microstructure

Published on: January 14, 2014

5.4K
Evaluation of Left Ventricular Structure and Function using 3D Echocardiography
06:34

Evaluation of Left Ventricular Structure and Function using 3D Echocardiography

Published on: October 28, 2020

3.9K

Area of Science:

  • Cardiology
  • Medical Imaging
  • Artificial Intelligence

Background:

  • Echocardiography-based assessment of left ventricular diastolic function (LVDF) according to ASE guidelines is complex and time-intensive.
  • Developing an automated approach is crucial for improving efficiency and consistency in clinical practice.

Purpose of the Study:

  • To develop a fully automated, lightweight hybrid algorithm combining deep learning (DL) and machine learning (ML) for LVDF assessment.
  • To enhance the speed and accuracy of diastolic function evaluation using echocardiographic data.

Main Methods:

  • A hybrid DL/ML algorithm (MMnet) was developed with multi-modality input and multi-task output.
  • The model measures LV ejection fraction (LVEF), left atrial end-systolic volume (LAESV), and key Doppler parameters (E, A, e', TRmax).
  • Training and testing were performed on internal datasets, with validation on three external datasets, including EchoNet-Dynamic and CAMUS.

Main Results:

  • MMnet achieved high segmentation accuracy (Dice 0.922-0.932) and classification accuracy (0.9977-1.0).
  • Mean absolute errors for LVEF and LAESV were low (3.7% and 5.8 ml, respectively), with external validation showing comparable results (4.9-5.6% for LVEF).
  • Diastolic function grading accuracy reached 0.88 with hard criteria and 0.98 with soft criteria.

Conclusions:

  • The MMnet algorithm automates the grading of ASE diastolic function with high accuracy and efficiency.
  • This automated approach utilizes 2D echocardiographic videos and Doppler images, offering a significant advancement in cardiac diagnostics.