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

Nonlinear hydrothermal associations between coupled landscape ecological risk and resilience in a major grain-producing region of China.

Journal of environmental management·2026

Same author

LangSurf: Language-Embedded Surface Gaussians for 3D Scene Understanding.

IEEE transactions on pattern analysis and machine intelligence·2026

Same author

A novel multi-task deep learning framework for classification and detection of intracranial structures in first-trimester fetal ultrasound images.

Physica medica : PM : an international journal devoted to the applications of physics to medicine and biology : official journal of the Italian Association of Biomedical Physics (AIFB)·2026

Same author

A new strategy for improving the precision and accuracy of isotope ratio analysis by quadrupole ICP-MS with the optimization of ion beam trajectory.

Analytical methods : advancing methods and applications·2026

Same author

Time-lapse ERT monitoring of the MPE-induced cone of depression at a low-permeability NAPL site: A field-based quantitative assessment of array-dependent geometric fidelity.

Journal of contaminant hydrology·2026

Same author

Molecular biology of targeting lymphangiogenesis in cancer: Therapeutic strategies and future promise.

Fundamental research·2026

Search research articles

Related Experiment Video

Updated: May 11, 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

Adversarial class-wise self-knowledge distillation for medical image segmentation.

Xiangchun Yu¹, Jiaqing Shen², Dingwen Zhang²

¹Jiangxi Provincial Key Laboratory of Multidimensional Intelligent Perception and Control, School of Information Engineering, Jiangxi University of Science and Technology, Ganzhou, China. yuxc@jxust.edu.cn.

Scientific Reports

|April 17, 2025

Summary

This summary is machine-generated.

This study introduces Adversarial Class-Wise Self-Knowledge Distillation (ACW-SKD) to improve medical image segmentation by reducing inter-class similarity. ACW-SKD enhances accuracy for difficult classes and offers efficient deployment on mobile devices.

Keywords:

Adversarial temperature loss Class-wise feature Inter-class similarity Medical image segmentation Self-knowledge distillation

More Related Videos

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

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

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

Published on: August 13, 2014

Related Experiment Videos

Last Updated: May 11, 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

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

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

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

Published on: August 13, 2014

Area of Science:

Medical Image Analysis
Computer Vision
Machine Learning

Background:

Medical image segmentation is crucial for diagnosis but challenged by inter-class similarity.
Existing knowledge distillation methods struggle to effectively address this challenge.
Optimizing compact student models requires effective learning objectives for improved performance.

Purpose of the Study:

To propose a novel self-knowledge distillation method, Adversarial Class-Wise Self-Knowledge Distillation (ACW-SKD), for medical image segmentation.
To mitigate the interference of inter-class similarity in segmentation tasks.
To develop an efficient model suitable for mobile device deployment.

Main Methods:

ACW-SKD utilizes an auxiliary head for coarse segmentation, refining class-wise features.
Class-wise feature distillation is employed to reduce inter-class similarity.
A feature reconstruction module (FRM) and adversarial temperature loss are incorporated for enhanced learning objectives.

Main Results:

ACW-SKD outperformed existing offline and self-knowledge distillation methods and U-Net on Synapse, FLARE2022, and M2caiSeg datasets.
The method significantly improved segmentation accuracy for challenging classes.
ACW-SKD demonstrated comparable performance to U-Net with reduced computational cost.

Conclusions:

ACW-SKD effectively addresses inter-class similarity in medical image segmentation.
The proposed method offers an efficient and accurate solution for medical image segmentation, suitable for mobile deployment.
ACW-SKD presents a promising advancement in knowledge distillation for medical imaging applications.