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

Accuracy of new-generation intraocular lens formulas in IOL power calculation for cataract extraction combined with vitrectomy.

International ophthalmology·2026
Same author

"Lotus leaf meets nacre": a dual biomimetic approach to creating highly humidity-stable films with exceptional barrier properties for food packaging.

Food chemistry·2026
Same author

A generalized two-step floating building catchment area model for measuring EMS accessibility with in-building patient access.

BMC health services research·2026
Same author

Determinants of Fear of Cancer Recurrence and Development of a Nomogram-Based Risk Prediction Model in Patients with Primary Liver Cancer: A Single-Center Cross-Sectional Study.

Journal of hepatocellular carcinoma·2026
Same author

A deep learning and radiomics fusion model enhances endoscopic ultrasonography diagnosis of gastric tumors.

Frontiers in oncology·2026
Same author

Elevated temperature enhances rpoE/degP-dependent bacterial membrane vesicle biogenesis and bla<sub>NDM-5</sub> dissemination in pig-derived carbapenem-resistant Escherichia coli.

Veterinary microbiology·2026

Related Experiment Video

Updated: Sep 3, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.9K

GREN: Graph-Regularized Embedding Network for Weakly-Supervised Disease Localization in X-Ray Images.

Baolian Qi, Gangming Zhao, Xin Wei

    IEEE Journal of Biomedical and Health Informatics
    |July 27, 2022
    PubMed
    Summary
    This summary is machine-generated.

    This study introduces a Graph Regularized Embedding Network (GREN) for more accurate disease localization in chest X-rays. GREN effectively uses relationships within and across images, improving upon existing weakly-supervised methods.

    More Related Videos

    A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports
    07:35

    A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports

    Published on: October 13, 2023

    1.7K
    Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
    07:15

    Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

    Published on: August 16, 2020

    6.9K

    Related Experiment Videos

    Last Updated: Sep 3, 2025

    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
    04:48

    Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

    Published on: November 30, 2022

    2.9K
    A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports
    07:35

    A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports

    Published on: October 13, 2023

    1.7K
    Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
    07:15

    Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

    Published on: August 16, 2020

    6.9K

    Area of Science:

    • Medical Imaging
    • Artificial Intelligence
    • Computer Vision

    Background:

    • Weakly-supervised disease localization in chest X-rays aims to reduce annotation effort.
    • Existing methods like multi-instance learning (MIL) and class activation maps (CAM) often produce inaccurate or incomplete localization regions.
    • These limitations stem from neglecting relationships between anatomical regions within an image and across different images.

    Purpose of the Study:

    • To develop a novel method for more accurate and integral disease localization in chest X-ray images.
    • To leverage both intra-image and inter-image relationships as contextual information for improved localization.
    • To address the shortcomings of current weakly-supervised approaches in capturing pathological implications.

    Main Methods:

    • Proposed the Graph Regularized Embedding Network (GREN) to model cross-region and cross-image relationships.
    • Utilized a pre-trained U-Net for lung lobe segmentation to establish intra-image anatomical context.
    • Employed intra-image and inter-image graphs, regularized by hash coding and Hamming distance, to capture structural information and facilitate training.

    Main Results:

    • Achieved state-of-the-art performance on the NIH chest X-ray dataset for weakly-supervised disease localization.
    • Demonstrated the effectiveness of GREN in generating more consistent and integral localization regions compared to prior methods.
    • Validated the importance of incorporating cross-region and cross-image relationships for accurate localization.

    Conclusions:

    • The proposed GREN effectively utilizes graph-based modeling of intra- and inter-image relationships for superior weakly-supervised disease localization.
    • This approach mimics a radiologist's comparative analysis, leading to enhanced diagnostic accuracy in chest X-rays.
    • The method offers a promising advancement in automated medical image analysis, with accessible code for further research.