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 11, 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

Multi-group deformable convolution network for 3D medical image segmentation.

Yuheng Li^1,2, Mingzhe Hu¹, Jing Wang³

¹Department of Radiation Oncology and Winship Cancer Institute, Emory University, Atlanta, Georgia, USA.

Medical Physics

|November 8, 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

A generalist biomedical vision-language model via multi-CLIP knowledge distillation.

Nature communications·2026

Same author

Transformer-based synthesis of 7 Tesla-like T1 contrast from routine clinical magnetic resonance imaging.

Medical physics·2026

Same author

Radiation Oncology-Biology Integration Network: Bridging the Gap between Biological Research and Clinical Practice.

Clinical cancer research : an official journal of the American Association for Cancer Research·2026

Same author

AURKA inhibitor VIC-1911 induces mitotic defects and functional BRCAness, sensitizing prostate cancer to PARP inhibition.

JCI insight·2026

Same author

Vision foundation model for 3D magnetic resonance imaging segmentation, classification, and registration.

Medical image analysis·2026

Same author

An efficient 3D latent diffusion model for T1-contrast enhanced MRI generation.

Biomedical physics & engineering express·2026

Same journal

Correction to "On the shape of the radiation survival curve in tumor spheroids: The role of oxygen heterogeneity".

Medical physics·2026

Same journal

Multi-view constrained semi-supervised vertebra detection for 3D ultrasound spine volume.

Medical physics·2026

Same journal

Accuracy of quantitative <sup>177</sup>Lu SPECT/CT imaging: A systematic review.

Medical physics·2026

Same journal

Physics-constrained dual-domain network for CBCT reconstruction from orthogonal X-rays in gynecologic radiotherapy.

Medical physics·2026

Same journal

Decomposition-based harmonization for quantitative PET imaging across scanners and radiotracers.

Medical physics·2026

Same journal

Development and evaluation of an in vivo dose-based monitoring system for electron FLASH radiation therapy.

Medical physics·2026

See all related articles

This summary is machine-generated.

MGDC-Net, a novel network for 3D medical image segmentation, effectively combines deformable convolutions and transformers. This approach achieves superior performance in segmenting brain tumors and organs, offering an efficient solution for radiation oncology.

Area of Science:

Medical image analysis
Computer vision
Machine learning

Background:

Accurate medical image segmentation is vital for radiation oncology, aiding in delineating anatomical structures and abnormalities for precise treatment planning.
Current 3D segmentation methods using Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) face challenges in capturing complex spatial and semantic information.

Purpose of the Study:

To develop an advanced 3D medical image segmentation network that overcomes the limitations of existing CNN and ViT approaches.
To enhance the capture of complex spatial and semantic structures in 3D medical images.

Main Methods:

Introduction of MGDC-Net, a Multi-Group Deformable Convolution network specifically designed for 3D volumetric medical image segmentation.
Integration of deformable convolution operators with learnable spatial offsets to focus on semantically important regions and transformer components to reduce CNN inductive biases.

Keywords:

computer tomography deformable convolution organ segmentation

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 11, 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

Leveraging stable spatial distribution across subjects for improved semantic learning and computational efficiency.

Main Results:

MGDC-Net achieved high Dice Similarity Coefficients (DSC) on diverse public datasets: 91.4% for brain tumor segmentation (BraTS21), 94.4% for CT multi-organ segmentation (FLARE21), and 84.1% for cross-modality MR/CT segmentation (AMOS22).
The network demonstrated superior segmentation performance across all evaluated tasks.
MGDC-Net also showed favorable computational efficiency compared to existing methods.

Conclusions:

MGDC-Net offers a robust and efficient solution for 3D volumetric medical image segmentation.
The network's combined use of deformable convolutions and transformer components shows significant potential for advancing medical image analysis applications.
The demonstrated improvements across multiple segmentation tasks highlight the method's versatility and effectiveness.