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

Clinical artificial intelligence applications of vision-language foundation models.

PLOS digital health·2026
Same author

Time-series single-cell transcriptomics and spatial metabolomics reveal spatiotemporal tobacco leaf response to herbivory.

Journal of advanced research·2026
Same author

A Knowledge-Guided Bi-Modal Network for the Classification of Anterior Chamber Angle Images.

IEEE journal of biomedical and health informatics·2026
Same author

Accelerating CEST MRI through complementary undersampling and multi-offset transformer reconstruction.

Communications engineering·2026
Same author

Enhancing SNR of CEST MRI Through Noise-to-Noise Deep Learning With K-Space Data Consistency.

NMR in biomedicine·2025
Same author

Data inversion of multi-dimensional magnetic resonance in porous media.

Magnetic resonance letters·2025

Related Experiment Video

Updated: Jul 8, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

406

A New Multi-Atlas Based Deep Learning Segmentation Framework With Differentiable Atlas Feature Warping.

Huabing Liu, Dong Nie, Jian Yang

    IEEE Journal of Biomedical and Health Informatics
    |December 19, 2023
    PubMed
    Summary

    This study introduces a novel deep learning framework for multi-atlas segmentation (DL-MA) that improves feature-level correspondence between atlases and target images, enhancing brain parcellation accuracy.

    More Related Videos

    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

    2.8K
    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
    14:08

    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

    Published on: April 13, 2013

    42.6K

    Related Experiment Videos

    Last Updated: Jul 8, 2025

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
    04:48

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    406
    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

    2.8K
    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
    14:08

    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

    Published on: April 13, 2013

    42.6K

    Area of Science:

    • Medical Image Analysis
    • Deep Learning
    • Neuroimaging

    Background:

    • Deep learning-based multi-atlas segmentation (DL-MA) methods excel in medical image segmentation, particularly brain parcellation.
    • Current DL-MA methods often struggle with atlas-target feature inconsistency due to image-level registration, limiting atlas information utility.
    • This inconsistency can negatively impact segmentation accuracy and the beneficial contribution of prior anatomical knowledge from atlases.

    Purpose of the Study:

    • To propose a novel DL-MA framework that addresses feature-level inconsistency for improved segmentation.
    • To introduce a differentiable atlas feature warping module with smooth regularization for enhanced atlas-target feature correspondence.
    • To achieve higher accuracy in medical image segmentation by making atlas features more relevant to target image features.

    Main Methods:

    • Development of a new deep learning-based multi-atlas segmentation framework.
    • Introduction of a differentiable atlas feature warping module designed for feature-level correspondence.
    • Inclusion of a novel smooth regularization term within the feature warping module.

    Main Results:

    • The proposed framework demonstrated statistically significant improvements over traditional multi-atlas segmentation (MAS) and existing DL-MA methods.
    • Evaluation on LPBA40 and NIREP-NA0 MR brain image datasets confirmed the framework's superior performance in brain parcellation.
    • Ablation studies validated the effectiveness and contribution of the differentiable atlas feature warping module.

    Conclusions:

    • The novel DL-MA framework effectively establishes feature-level correspondence, enhancing segmentation accuracy.
    • The differentiable atlas feature warping module is a key component for improving the relevance of atlas features to target images.
    • This approach offers a significant advancement for accurate brain parcellation and potentially other medical image segmentation tasks.