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

Knowledge-Based Feature Selection Substantially Enhances Data-Driven Wastewater Treatment Modeling.

Environmental science & technology·2026
Same author

Active-ester-based post-functionalized nanofibers with UV-switchable colorimetric response toward HCl and NH<sub>3</sub> vapors.

Nanoscale·2026
Same author

Advances in oral delivery systems for alpha-linolenic acid: mechanisms, bioavailability, and applications in functional foods.

Critical reviews in food science and nutrition·2026
Same author

Configuration-Dependent Nanoarchitectonic Effects of Linear and Cyclic Calcium Polyphosphates on Mitochondrial Energy Metabolism and Osteogenic Differentiation.

ACS applied materials & interfaces·2026
Same author

The "metamorphosis" of alum sludge for phosphate adsorption: Deciphering the interlayer microenvironment construction and molecular mechanism of intercalated layered double hydroxides.

Environmental research·2026
Same author

Towards the construction of a virtual yeast.

Nature·2026

Related Experiment Video

Updated: Jun 24, 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

2.7K

Prototype Correlation Matching and Class-Relation Reasoning for Few-Shot Medical Image Segmentation.

Yumin Zhang, Hongliu Li, Yajun Gao

    IEEE Transactions on Medical Imaging
    |June 11, 2024
    PubMed
    Summary

    This study introduces a new model for few-shot medical image segmentation that improves accuracy by matching prototypes and reasoning class relationships. It effectively handles variations within medical classes for better generalization to unseen classes.

    More Related Videos

    Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
    04:25

    Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

    Published on: December 15, 2023

    2.3K
    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

    48.0K

    Related Experiment Videos

    Last Updated: Jun 24, 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

    2.7K
    Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
    04:25

    Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

    Published on: December 15, 2023

    2.3K
    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

    48.0K

    Area of Science:

    • Biomedical Imaging
    • Computer Vision
    • Artificial Intelligence

    Background:

    • Few-shot medical image segmentation advances medical analysis accuracy and efficiency.
    • Existing methods struggle with inter-class relations and intra-class variations, hindering generalization to novel classes.
    • Large intra-class variations in medical images lead to ambiguous characterization and degraded performance.

    Purpose of the Study:

    • To propose a novel model, Prototype correlation Matching and Class-relation Reasoning (PMCR), to address challenges in few-shot medical image segmentation.
    • To mitigate false pixel correlation matches caused by large intra-class variations.
    • To reason inter-class relations among base and novel medical classes for improved segmentation of unseen classes.

    Main Methods:

    • Developed a prototype correlation matching module to identify representative prototypes and characterize diverse visual information.
    • Employed optimal transport algorithm for prototype-level correlation matching between support and query features, overcoming pixel-level limitations.
    • Designed a class-relation reasoning module to leverage inter-class relationships for segmenting novel medical objects.

    Main Results:

    • The PMCR model effectively mitigates false pixel correlation matches arising from significant intra-class variations.
    • The model successfully reasons inter-class relations, enhancing the semantic encoding of local query features.
    • Quantitative comparisons demonstrate substantial performance improvements of the proposed model over existing baseline methods.

    Conclusions:

    • The PMCR model offers a robust solution for few-shot medical image segmentation by effectively handling intra-class variations and leveraging inter-class reasoning.
    • The prototype-level matching and class-relation reasoning approach significantly improves generalization to unseen medical classes.
    • This work contributes to more accurate and efficient medical image analysis through advanced deep learning techniques.