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

Performance Gaps and Optimization Strategies in Chinese Medical Large Language Models Based on MedBench: Evaluation Study.

JMIR AI·2026
Same author

Case Report: Medical management of a rare infectious basilar artery aneurysm.

Frontiers in medicine·2026
Same author

PhotIQA: A photoacoustic image data set with image quality ratings.

Scientific data·2026
Same author

Advanced radiotherapy and systemic therapy in head and neck adenoid cystic carcinoma: Current progress and future integrated strategies.

Translational oncology·2026
Same author

Chemical Environments Influence DNA Nanostructures: From Fundamental Interactions to Functional Applications.

Chemistry (Weinheim an der Bergstrasse, Germany)·2026
Same author

A triaxial benchmark for assessing responses from large language models in traditional Chinese medicine.

Communications medicine·2026

Related Experiment Video

Updated: Jul 10, 2025

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

48.1K

Contrastive Registration for Unsupervised Medical Image Segmentation.

Lihao Liu, Angelica I Aviles-Rivero, Carola-Bibiane Schonlieb

    IEEE Transactions on Neural Networks and Learning Systems
    |November 20, 2023
    PubMed
    Summary

    This study introduces CLMorph, a novel unsupervised medical image segmentation method using contrastive learning and registration. CLMorph significantly improves segmentation accuracy, overcoming limitations of previous unsupervised approaches.

    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
    Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
    07:13

    Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

    Published on: October 27, 2023

    1.2K

    Related Experiment Videos

    Last Updated: Jul 10, 2025

    Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
    10:25

    Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

    Published on: September 25, 2019

    48.1K
    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
    Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
    07:13

    Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

    Published on: October 27, 2023

    1.2K

    Area of Science:

    • Medical Imaging
    • Artificial Intelligence
    • Computer Vision

    Background:

    • Supervised deep learning methods achieve high accuracy in medical image segmentation but require extensive labeled data, which is costly and time-consuming to obtain.
    • Existing unsupervised segmentation techniques lack the accuracy to compete with supervised methods, limiting their clinical applicability.
    • Human bias in manual annotations further complicates the creation of large, representative labeled datasets for medical imaging.

    Purpose of the Study:

    • To develop a novel unsupervised medical image segmentation technique that overcomes the limitations of existing methods.
    • To introduce a new convolutional neural network (CNN)-based contrastive registration architecture for accurate unsupervised segmentation.
    • To leverage image-level registration and feature-level contrastive learning for improved segmentation performance.

    Main Methods:

    • Proposed a novel optimization model integrated into a CNN-based contrastive registration architecture named CLMorph.
    • Developed an architecture to capture image-to-image transformation mappings through registration for unsupervised segmentation.
    • Embedded a contrastive learning mechanism within the registration architecture to enhance feature discrimination.

    Main Results:

    • The CLMorph technique effectively mitigates drawbacks associated with current unsupervised segmentation methods.
    • Demonstrated substantial performance improvements over state-of-the-art unsupervised segmentation techniques through numerical and visual experiments.
    • Achieved superior results on two major medical image datasets, validating the proposed approach.

    Conclusions:

    • The CLMorph approach offers a promising solution for accurate unsupervised medical image segmentation.
    • The combination of contrastive learning and registration in CLMorph enhances segmentation capabilities without requiring labeled data.
    • This method has the potential to reduce the reliance on manual annotations in clinical diagnosis and treatment planning.