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

Search research articles

Related Experiment Video

Updated: Jan 14, 2026

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

Uncertainty-aware collaborative learning with mixed images for semi-supervised medical image segmentation.

Meixi Wang¹, Xiumei Li¹, Huang Bai¹

¹School of Information Science and Technology, Hangzhou Normal University, Hangzhou, China.

Computer Methods and Programs in Biomedicine

|October 25, 2025

Summary

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

Study on the anti-inflammatory effects and mechanisms of gentisic acid based on the LPS-induced RAW264.7 cell inflammation model and the oxazolone-induced zebrafish inflammation model.

Frontiers in pharmacology·2026

Same author

Chitosan-based nanotheranostics integrated dual-modal imaging and combinatorial tumor therapy for EGFR-TKI resistance reversal.

Materials today. Bio·2026

Same author

X-ray irradiation simultaneously mitigates chlorfenapyr and azoxystrobin residues and preserves postharvest quality of cowpea.

Food chemistry: X·2026

Same author

Optimizing cutoffs for clinical interpretation of brain amyloid status using PET/MRI: a multisite study.

Alzheimer's research & therapy·2026

Same author

DNA-PK-mediated CRTC2 phosphorylation promotes NHEJ and suppresses antitumor immunity via relocation to repair complexes.

Nature communications·2026

Same author

Caffeic Acid-Chicoric Acid (1:1) Mixture Exerts Antioxidant Effects and Regulates Intestinal Health in Oxazolone-Induced Zebrafish.

Antioxidants (Basel, Switzerland)·2026

This study introduces a new framework for semi-supervised medical image segmentation, improving accuracy by reducing noisy pseudo-labels and leveraging diverse data perturbations for better generalization in medical AI.

Area of Science:

Medical Image Analysis
Artificial Intelligence in Medicine
Computer Vision

Background:

Semi-supervised medical image segmentation (SSMIS) uses unlabeled data to reduce annotation costs.
Pseudo-labeling (PL) is common in SSMIS but struggles with noisy labels from limited annotations.
Existing methods using single perturbations have limited generalization; multiple perturbations pose training challenges.

Purpose of the Study:

To address the challenges of noisy pseudo-labels and limited generalization in PL-based SSMIS.
To improve the accuracy and robustness of medical image segmentation models.
To develop a framework that effectively utilizes both labeled and unlabeled data.

Main Methods:

Proposed an Uncertainty-Aware Collaborative Learning with Mixed Images (CLMI) framework for SSMIS.

Keywords:

Medical image segmentation Prototype learning Pseudo-labels Semi-supervised learning Uncertainty estimation

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Related Experiment Videos

Last Updated: Jan 14, 2026

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Introduced Mixed Prototype guided Uncertainty Estimation (MPUE) to reduce pseudo-label noise by estimating uncertainty.

CLMI utilizes both perturbed and unperturbed images to enhance generalization and mitigate overfitting.

Main Results:

Achieved Dice scores of 89.97% on ACDC and 81.20% on PROMISE12 datasets with only 10% labeled data.
Demonstrated significant outperformance compared to state-of-the-art semi-supervised methods.
Validated the framework's effectiveness on public medical image datasets.

Conclusions:

The proposed framework enhances the accuracy of semi-supervised medical image segmentation.
Experiments confirm the framework's superiority over existing methods.
Code is publicly available for reproducibility and further research.