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

Correction: A multi-center, randomized, double-blind, placebo-controlled trial to evaluate the efficacy and safety of Fuzheng Yangxin Granule in treating heart failure with preserved ejection fraction (Qi-Yin deficiency and blood stasis syndrome): study protocol.

Frontiers in cardiovascular medicine·2026
Same author

Efficacy and safety of Salvia miltiorrhiza and ligustrazine injection for heart failure: a systematic review and meta-analysis.

Frontiers in pharmacology·2026
Same author

A multi-center, randomized, double-blind, placebo-controlled trial to evaluate the efficacy and safety of Fuzheng Yangxin Granule in treating heart failure with preserved ejection fraction (Qi-Yin deficiency and blood stasis syndrome): study protocol.

Frontiers in cardiovascular medicine·2025
Same author

Spatial and temporal changes of ecosystem service value and its influencing mechanism in the Yangtze River Delta urban agglomeration.

Scientific reports·2024
Same author

Liver impairment in COVID-19 patients: A retrospective analysis of 115 cases from a single centre in Wuhan city, China.

Liver international : official journal of the International Association for the Study of the Liver·2020
Same author

Role of oral microbiota in atherosclerosis.

Clinica chimica acta; international journal of clinical chemistry·2020

Related Experiment Video

Updated: Jul 18, 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

2.8K

Poisson-based image editing for semi-supervised vitiligo lesion segmentation with limited annotations.

Jiacong Wang1, Xiaolan Ding2, Jun Xiao1

  • 1School of Artificial Intelligence, University of Chinese Academy of Sciences, China.

Computers in Biology and Medicine
|August 25, 2023
PubMed
Summary

This study introduces an image editing data augmentation strategy and a semi-supervised approach to improve vitiligo lesion segmentation, addressing challenges in medical image analysis with limited annotated data.

Keywords:
Data augmentationMedical image segmentationPseudo-labelsSemi-supervised learning

More Related Videos

A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment
10:39

A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment

Published on: May 24, 2022

2.4K
From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

24.6K

Related Experiment Videos

Last Updated: Jul 18, 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

2.8K
A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment
10:39

A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment

Published on: May 24, 2022

2.4K
From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

24.6K

Area of Science:

  • Medical Image Analysis
  • Computer Vision
  • Dermatology

Background:

  • Vitiligo lesion segmentation is vital for treatment assessment but faces challenges due to limited annotated data and image collection.
  • Existing methods often require large labeled datasets, hindering reproducibility and transferability in data-scarce domains.
  • There is a need for innovative approaches that utilize unlabeled data for segmentation with minimal annotated samples.

Purpose of the Study:

  • To develop a data augmentation strategy for synthesizing high-quality vitiligo images from limited annotated data.
  • To adapt the Mean-Teacher framework for effective semi-supervised learning, reducing reliance on dense annotations.
  • To introduce the Bimodal Vitiligo Lesions Segmentation (BVLS) dataset to address the scarcity of vitiligo segmentation datasets.

Main Methods:

  • A novel image editing-based data augmentation technique to generate diverse and visually realistic training samples.
  • Implementation of the Mean-Teacher framework to mine reliable predictions from unlabeled data using high-confidence pseudo-labels.
  • Creation and utilization of the Bimodal Vitiligo Lesions Segmentation (BVLS) dataset, featuring detailed segmentation masks and bimodal images.

Main Results:

  • The proposed data augmentation strategy significantly improved segmentation performance (+17.27%) on the UNet backbone compared to existing methods.
  • The semi-supervised framework achieved a notable Intersection over Union (IoU) of 49.71% using only 10% of annotated images.
  • The developed BVLS dataset and augmentation techniques enhance the segmentation accuracy for vitiligo lesions.

Conclusions:

  • The combined approach of image editing data augmentation and semi-supervised learning effectively addresses the limitations of data scarcity in vitiligo segmentation.
  • The method offers a cost-effective solution for improving segmentation performance without requiring extensive manual annotation.
  • The study provides a valuable resource (BVLS dataset and code) for advancing research in automated vitiligo assessment.