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: Nov 15, 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

Deep virtual adversarial self-training with consistency regularization for semi-supervised medical image

Xi Wang¹, Hao Chen¹, Huiling Xiang²

¹Department of Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong, China.

Medical Image Analysis

|March 7, 2021

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

M phase phosphorylation of the epigenetic regulator UHRF1 regulates its physical association with the deubiquitylase USP7 and stability.

Proceedings of the National Academy of Sciences of the United States of America·2012

Same author

Biosynthesis of ethyl oleate, a primer pheromone, in the honey bee (Apis mellifera L.).

Insect biochemistry and molecular biology·2012

Same author

Co-delivery strategies based on multifunctional nanocarriers for cancer therapy.

Current drug metabolism·2012

Same author

Efficacy of gemifloxacin for the treatment of experimental Staphylococcus aureus keratitis.

Journal of ocular pharmacology and therapeutics : the official journal of the Association for Ocular Pharmacology and Therapeutics·2012

Same author

Expression profile analysis of the polygalacturonase-inhibiting protein genes in rice and their responses to phytohormones and fungal infection.

Plant cell reports·2012

Same author

Characterizing natural dissolved organic matter in a freshly submerged catchment (Three Gorges Dam, China) using UV absorption, fluorescence spectroscopy and PARAFAC.

Water science and technology : a journal of the International Association on Water Pollution Research·2012

Same journal

Wavelet-inspired diffusion model with near-field constraint for real-time echocardiography dehazing.

Medical image analysis·2026

Same journal

Co-assistant networks by pathology foundation model and convolutional neural network for gigapixel whole slide image analysis.

Medical image analysis·2026

Same journal

MBAS2024: A large-scale benchmark for multi-class bi-atrial segmentation in multi-center contrast-enhanced MRIs.

Medical image analysis·2026

Same journal

Respiratory motion augmentation for personalized super-resolution (RMApSR) of 3D cine MR images in MRI-guided radiotherapy.

Medical image analysis·2026

Same journal

Biom3d, a modular framework to host and develop 3D segmentation methods.

Medical image analysis·2026

Same journal

Embracing intra-class heterogeneity for semi-supervised medical image segmentation: From diversity to precision.

Medical image analysis·2026

See all related articles

This summary is machine-generated.

This study introduces a new semi-supervised deep learning method for medical image classification, significantly improving accuracy by utilizing unlabeled data through self-training and consistency regularization. The novel approach enhances diagnostic capabilities in challenging medical imaging tasks.

Area of Science:

Medical Imaging
Artificial Intelligence
Computer Vision

Background:

Supervised deep learning excels in medical imaging but requires extensive labeled data, which is costly and time-consuming to acquire.
The scarcity of labeled medical data presents a significant challenge for traditional supervised learning models.

Purpose of the Study:

To develop a novel semi-supervised deep learning method for large-scale medical image classification.
To effectively leverage unlabeled medical data to improve model performance and reduce reliance on expert annotations.

Main Methods:

Proposed a semi-supervised deep learning method: deep virtual adversarial self-training with consistency regularization.
Utilized self-training for pseudo-labeling high-confidence predictions on weakly-augmented data.

Keywords:

Consistency regularization Deep learning Semi-supervised classification

Related Experiment Videos

Last Updated: Nov 15, 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

Implemented consistency regularization by enforcing agreement between predictions on weakly- and strongly-augmented data.

Incorporated virtual adversarial training (VAT) on both labeled and unlabeled data to enhance robustness.

Main Results:

The proposed method significantly outperformed supervised baselines and state-of-the-art methods on medical image classification tasks.
Demonstrated superior performance in breast cancer screening from ultrasound images.
Achieved high accuracy in multi-class ophthalmic disease classification from optical coherence tomography (OCT) B-scan images.

Conclusions:

The novel semi-supervised deep learning approach effectively utilizes unlabeled data to enhance medical image classification.
The method shows great promise for improving diagnostic accuracy in resource-limited settings.
This work offers a robust solution for medical image analysis challenges where labeled data is scarce.