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

SARS-CoV-2 infection and COVID-19 vaccination and the risk for new-onset type 1 diabetes: a register-based population study in Sweden.

Diabetologia·2026
Same author

Mental illness after bereavement before and during the COVID-19 pandemic in Sweden: A matched cohort study.

PLOS mental health·2026
Same author

Bereavement and Risk of Cardiovascular Disease Before and During the COVID-19 Pandemic.

JAMA network open·2026
Same author

Ursodeoxycholic acid treatment did not show protective effect for severe COVID-19 outcomes - a nationwide register study.

BMC public health·2026
Same author

Cardiovascular and mortality risks following COVID-19-related vs. non-COVID-19 COPD exacerbations.

Annals of the American Thoracic Society·2026
Same author

Taming Proton Transfer Through Proton Conductors to Boost Hydrogen Evolution.

Angewandte Chemie (International ed. in English)·2026

Related Experiment Video

Updated: Jul 12, 2025

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.1K

Encoder-Decoder Contrast for Unsupervised Anomaly Detection in Medical Images.

Jia Guo, Shuai Lu, Lize Jia

    IEEE Transactions on Medical Imaging
    |October 26, 2023
    PubMed
    Summary

    This study introduces Encoder-Decoder Contrast (EDC), a new unsupervised anomaly detection method for medical images. EDC optimizes the entire network to improve accuracy in detecting anomalies in medical scans.

    More Related Videos

    Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
    05:56

    Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

    Published on: April 14, 2023

    2.5K
    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

    Related Experiment Videos

    Last Updated: Jul 12, 2025

    Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
    09:47

    Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

    Published on: December 15, 2023

    1.1K
    Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
    05:56

    Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

    Published on: April 14, 2023

    2.5K
    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

    Area of Science:

    • Medical Image Analysis
    • Computer Vision
    • Machine Learning

    Background:

    • Unsupervised anomaly detection (UAD) in medical imaging leverages normal images to identify abnormalities, avoiding costly labeling of unhealthy samples.
    • Current UAD methods often use pre-trained encoders on natural images, leading to domain mismatch and pattern collapse issues in medical applications.
    • Optimizing encoders can cause pattern collapse, limiting the effectiveness of existing UAD techniques.

    Purpose of the Study:

    • To propose a novel unsupervised anomaly detection (UAD) method, Encoder-Decoder Contrast (EDC), for medical image analysis.
    • To address the limitations of domain-specific feature extraction and pattern collapse in current UAD methods.
    • To develop a method that optimizes the entire network for target medical domains, enhancing anomaly detection performance.

    Main Methods:

    • Introduced Encoder-Decoder Contrast (EDC), a UAD method optimizing the entire network, including encoder and decoder, for medical image analysis.
    • Integrated a contrastive learning paradigm to mitigate pattern collapse during simultaneous encoder and decoder optimization.
    • Incorporated globality into the contrastive objective function to improve stability and performance.

    Main Results:

    • The proposed EDC method demonstrated superior performance across four diverse medical imaging modalities: optical coherence tomography, color fundus images, brain MRI, and skin lesion images.
    • EDC outperformed all current state-of-the-art unsupervised anomaly detection methods in conducted experiments.
    • The method effectively reduced biases from pre-trained natural image domains and oriented the network towards target medical image features.

    Conclusions:

    • Encoder-Decoder Contrast (EDC) offers a robust and effective solution for unsupervised anomaly detection in medical imaging.
    • The proposed method overcomes key limitations of existing UAD techniques, particularly domain mismatch and pattern collapse.
    • EDC provides a promising advancement for automated analysis and diagnosis in various medical imaging applications.