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 Video

Updated: Jul 16, 2026

3D Whole-heart Myocardial Tissue Analysis
06:53

3D Whole-heart Myocardial Tissue Analysis

Published on: April 12, 2017

A Prototype-Guided 3D Deep Learning Framework for Myocardial Perfusion Scintigraphy Segmentation.

Madallah Alruwaili1, Mahmood A Mahmood2

  • 1Department of Computer Engineering and Networks, College of Computer and Information Sciences, Jouf University, Sakaka 72388, Saudi Arabia.

Journal of Clinical Medicine
|July 15, 2026
PubMed
Summary

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

An Explainable Multi-Scale Deep Learning Framework for Multi-Class Brain MRI Classification.

Diagnostics (Basel, Switzerland)·2026
Same author

A Dual-Branch Frequency-Aware Attention Framework for Rare Neurological Disease Classification from Brain MRI.

Diagnostics (Basel, Switzerland)·2026
Same author

Progression-Aware and Explainable CNN-Transformer Framework for Multiclass Alzheimer's Disease Staging Using MRI.

Diagnostics (Basel, Switzerland)·2026
Same author

Dual-Attention EfficientNet Hybrid U-Net for Segmentation of Rheumatoid Arthritis Hand X-Rays.

Diagnostics (Basel, Switzerland)·2025
Same author

Decoding Multi-Omics Signatures in Lower-Grade Glioma Using Protein-Protein Interaction-Informed Graph Attention Networks and Ensemble Learning.

Diagnostics (Basel, Switzerland)·2025
Same author

FracFusionNet: A Multi-Level Feature Fusion Convolutional Network for Bone Fracture Detection in Radiographic Images.

Diagnostics (Basel, Switzerland)·2025

CardioProto-SegNet offers a new deep learning framework for segmenting myocardial regions in myocardial perfusion scintigraphy (MPS) images. This method establishes a reproducible benchmark for future quantitative analysis of coronary artery disease.

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Cardiovascular Disease Research

Background:

  • Myocardial perfusion scintigraphy (MPS) is crucial for assessing coronary artery disease (CAD).
  • Limited availability of public datasets hinders reproducible deep learning studies in MPS.
  • This study addresses the need for a standardized dataset and segmentation framework.

Purpose of the Study:

  • To introduce CardioProto-SegNet, a 3D deep learning framework for myocardial segmentation.
  • To utilize the public PhysioNet Myocardial Perfusion Scintigraphy Image Database for model development and validation.
  • To establish a reproducible benchmark for image-only myocardium segmentation in MPS.

Main Methods:

  • Developed an image-only 3D anatomy-directed segmentation framework (CardioProto-SegNet).
Keywords:
3D deep learningSPECTanatomy-aware learningmedical image segmentationmyocardial perfusion imagingmyocardial perfusion scintigraphyprototype memory refinementsingle-photon emission computed tomography

Related Experiment Videos

Last Updated: Jul 16, 2026

3D Whole-heart Myocardial Tissue Analysis
06:53

3D Whole-heart Myocardial Tissue Analysis

Published on: April 12, 2017

  • Employed a U-Net-like residual encoder-decoder network with attention mechanisms.
  • Refined the model using prototype-memory at the bottleneck for enhanced segmentation.

    • Main Results:

    • CardioProto-SegNet achieved a Dice score of 0.7402 on the holdout test set.
    • Cross-validation yielded a mean Dice score of 0.8239 and mean IoU of 0.6870.
    • Ablation studies confirmed the importance of residual connections and identified optimal model depth.

    Conclusions:

    • CardioProto-SegNet provides a robust and reproducible method for myocardium segmentation in MPS.
    • The framework serves as a foundation for future quantitative analysis and CAD research.
    • Future work will leverage larger datasets with clinical labels for enhanced downstream analysis.