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: Sep 21, 2025

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

2.0K

Augmented Equivariant Attention Networks for Microscopy Image Transformation.

Yaochen Xie, Yu Ding, Shuiwang Ji

    IEEE Transactions on Medical Imaging
    |June 1, 2022
    PubMed
    Summary

    Deep learning models can now create high-quality microscopy images from low-quality ones. New augmented equivariant attention networks (AEANets) improve image transformation by better capturing dependencies between images.

    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

    Regulating Electron-Density Distribution of Pyridine Derivative Passivators for Efficient Carbon-Based CsPbI<sub>3-x</sub>Br<sub><i>x</i></sub> Perovskite Solar Cells.

    The journal of physical chemistry letters·2026
    Same author

    Augmenting Molecular Graphs with Geometries via Machine Learning Interatomic Potentials.

    ArXiv·2025
    Same author

    Resveratrol ameliorates Salmonella Typhimurium-induced intestinal inflammation and barrier dysfunction in chickens via COX-2 inhibition.

    Research in veterinary science·2025
    Same author

    A Benchmark for Quantum Chemistry Relaxations via Machine Learning Interatomic Potentials.

    ArXiv·2025
    Same author

    Correction: DiffBP: generative diffusion of 3D molecules for target protein binding.

    Chemical science·2025
    Same author

    Learning to Discover Regulatory Elements for Gene Expression Prediction.

    ArXiv·2025

    Area of Science:

    • Microscopy imaging
    • Deep learning
    • Computational imaging

    Background:

    • High-quality electron microscopy (EM) and fluorescence microscopy (FM) imaging is costly, time-consuming, and can damage samples.
    • Deep learning offers computational solutions for microscopy image enhancement tasks like denoising and super-resolution.
    • Previous models struggle with inter-image dependencies and preserving spatial equivariance, limiting performance in image-to-image transformations.

    Purpose of the Study:

    • To develop advanced deep learning models for microscopy image transformation.
    • To address limitations in existing models regarding inter-image dependency and equivariance preservation.
    • To introduce a novel network architecture capable of enhanced image-to-image translation in microscopy.

    Main Methods:

    More Related Videos

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

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    530

    Related Experiment Videos

    Last Updated: Sep 21, 2025

    A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
    04:23

    A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

    Published on: April 21, 2023

    2.0K
    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
    04:48

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    530
    • Proposed Augmented Equivariant Attention Networks (AEANets) to capture inter-image dependencies and shared features.
    • Implemented two attention mechanism augmentations: shared references and batch-aware attention.
    • Theoretically derived and experimentally validated the equivariance property of the proposed model.

    Main Results:

    • AEANets demonstrated superior performance in capturing inter-image dependencies and shared features compared to baseline methods.
    • The proposed model consistently outperformed existing methods in both quantitative metrics and visual quality.
    • The equivariance property was successfully preserved, crucial for accurate image transformation.

    Conclusions:

    • AEANets offer a significant advancement in computational microscopy image enhancement.
    • The novel approach effectively addresses the limitations of prior deep learning models for image-to-image tasks.
    • This work paves the way for more efficient and less invasive high-quality microscopy imaging.