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

Few-mode fiber based Raman distributed temperature sensing.

Optics express·2017
Same author

Tripodal S-Ligand Complexes of Copper(I) as Catalysts for Alkene Aziridination, Sulfide Sulfimidation, and C-H Amination.

Inorganic chemistry·2017
Same author

5-aminolevulinic acid-mediated photodynamic therapy and its strain-dependent combined effect with antibiotics on Staphylococcus aureus biofilm.

PloS one·2017
Same author

Optimized Multiresidue Analysis of Organic Contaminants of Priority Concern in a Daily Consumed Fish (Grass Carp).

Journal of analytical methods in chemistry·2017
Same author

Deubiquitinating enzyme USP20 is a positive regulator of Claspin and suppresses the malignant characteristics of gastric cancer cells.

International journal of oncology·2017
Same author

Evaluation of the implementation rate of primary antifungal prophylaxis and the prognosis of invasive fungal disease in acute leukemia patients in China.

Journal of infection and chemotherapy : official journal of the Japan Society of Chemotherapy·2017

Related Experiment Video

Updated: Nov 24, 2025

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

43.1K

Learn to Threshold: ThresholdNet With Confidence-Guided Manifold Mixup for Polyp Segmentation.

Xiaoqing Guo, Chen Yang, Yajie Liu

    IEEE Transactions on Medical Imaging
    |December 28, 2020
    PubMed
    Summary

    This study introduces ThresholdNet, a new deep learning model that improves polyp segmentation in endoscopy images using a novel data augmentation technique called CGMMix. The method enhances early colorectal cancer diagnosis by overcoming data limitations and class imbalance issues.

    More Related Videos

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
    04:48

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    655
    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

    3.1K

    Related Experiment Videos

    Last Updated: Nov 24, 2025

    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
    14:08

    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

    Published on: April 13, 2013

    43.1K
    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
    04:48

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    655
    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

    3.1K

    Area of Science:

    • Medical Imaging
    • Computer Vision
    • Artificial Intelligence

    Background:

    • Automatic polyp segmentation in endoscopy is vital for early colorectal cancer detection.
    • Current deep learning methods struggle with limited annotated data and class imbalance.
    • Existing segmentation methods use a fixed threshold, limiting accuracy.

    Purpose of the Study:

    • To develop an improved deep learning model for polyp segmentation in endoscopy images.
    • To address challenges of limited datasets and class imbalance in polyp segmentation.
    • To enhance the accuracy and reliability of polyp detection for early cancer diagnosis.

    Main Methods:

    • Proposed a novel ThresholdNet architecture with a confidence-guided manifold mixup (CGMMix) data augmentation technique.
    • CGMMix performs mixup at image and feature levels, guided by confidence to address class imbalance.
    • Introduced two consistency regularizations (MFMC and MCMC) and a threshold map supervision generator (TMSG).

    Main Results:

    • Achieved state-of-the-art performance on two polyp segmentation datasets (EndoScene and WCE polyp).
    • Demonstrated superior dice scores of 87.307% on EndoScene and 87.879% on WCE polyp dataset.
    • The ThresholdNet effectively calibrates segmentation results using the learned threshold map.

    Conclusions:

    • The proposed ThresholdNet and CGMMix method significantly advance polyp segmentation accuracy.
    • This approach effectively mitigates issues of limited data and class imbalance in medical image analysis.
    • The developed method holds promise for improving early diagnosis and treatment of colorectal cancer.