Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

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

Proteomics and human microchips identify Thrombospondin-1 as a potential biomarker for calciphylaxis stem cell therapy.

iScience·2026

Same author

Influence Mechanism of N<sub>2</sub>-CO<sub>2</sub> Mixtures on the CH<sub>4</sub> Displacement Behavior from Different Coal Samples under Stress Condition.

ACS omega·2026

Same author

Efficient and accurate neural-field reconstruction using resistive memory.

Nature·2026

Same author

Integrated metabolomics and transcriptomics reveal metabolite differences between wild and cultivated Angelica sinensis.

BMC plant biology·2026

Same author

Afternoon Anesthesia Induction is Associated with Post-Induction Hypotension in Patients Undergoing Off-Pump Coronary Artery Bypass Grafting: A Retrospective, Single-Centre Study Using Propensity Score Matching

Journal of cardiothoracic and vascular anesthesia·2026

Same author

<i>Akkermansia muciniphila</i>-derived L-norleucine modulates FABP1-dependent fatty acid transport.

Proceedings of the National Academy of Sciences of the United States of America·2026

Search research articles

Related Experiment Video

Updated: Jul 29, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Compete to Win: Enhancing Pseudo Labels for Barely-Supervised Medical Image Segmentation.

Huimin Wu, Xiaomeng Li, Yiqun Lin

IEEE Transactions on Medical Imaging

|May 23, 2023

Summary

This study introduces Compete-to-Win (ComWin), a novel method for barely-supervised medical image segmentation. ComWin enhances pseudo label quality by comparing multiple network predictions, improving segmentation accuracy with limited data.

More Related Videos

Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench

Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench

Published on: August 23, 2017

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

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

Published on: April 14, 2023

Related Experiment Videos

Last Updated: Jul 29, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench

Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench

Published on: August 23, 2017

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

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

Published on: April 14, 2023

Area of Science:

Medical Image Analysis
Computer Vision
Machine Learning

Background:

Barely-supervised medical image segmentation faces challenges with limited labeled data.
Existing semi-supervised methods like cross pseudo supervision struggle with foreground class precision.

Purpose of the Study:

To propose a novel method, Compete-to-Win (ComWin), to enhance pseudo label quality in barely-supervised medical image segmentation.
To improve the precision of foreground class segmentation under data scarcity.

Main Methods:

ComWin generates high-quality pseudo labels by comparing confidence maps from multiple networks, selecting the most confident prediction.
ComWin+ integrates a boundary-aware enhancement module for refined pseudo labels near object boundaries.

Main Results:

ComWin and ComWin+ achieve state-of-the-art performance on three public medical image datasets.
The method demonstrates superior results in cardiac structure, pancreas, and colon tumor segmentation.

Conclusions:

The Compete-to-Win strategy effectively enhances pseudo label quality for barely-supervised medical image segmentation.
ComWin and its enhanced version offer a robust solution for segmentation tasks with limited labeled data.