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

TREM1 signaling amplifies neutrophil-mediated inflammation in inflammatory bowel disease.

Nature communications·2026
Same author

Management of Engraftment Arrhythmias Associated with Human Induced Pluripotent Stem Cell-Derived Cardiomyocytes Transplantation.

Heart rhythm·2026
Same author

Herkinorin Exerted Neuroprotective Effects After Spinal Cord Injury by Suppression of NF-κB Pathway Activation.

Neurochemical research·2026
Same author

Neural Network-Based Prediction of Residual Paravalvular Leak in Bicuspid Aortic Valve TAVI Using CT-Derived Anatomical Features.

Biomedicines·2026
Same author

Correlations of pterygial blood flow with corneal refractive properties and higher-order aberrations: A cross-sectional OCTA study.

Photodiagnosis and photodynamic therapy·2026
Same author

Rapid endothelialization induced by VEGF-mimetic peptide in small-diameter vascular scaffold.

Biomedical materials (Bristol, England)·2026

Related Experiment Video

Updated: Aug 23, 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

2.9K

Mixture 2D Convolutions for 3D Medical Image Segmentation.

Jianyong Wang1, Lei Zhang1, Yi Zhang1

  • 1Machine Intelligence Laboratory, College of Computer Science, Sichuan University, Chengdu, Sichuan, P. R. China.

International Journal of Neural Systems
|November 3, 2022
PubMed
Summary

A new Mixture Convolutional Network (MixConvNet) improves 3D medical image segmentation by efficiently combining 2D and 3D convolutions. This novel approach enhances accuracy while reducing computational costs for better medical imaging analysis.

Keywords:
Mixture convolutional networkdeep neural networkmedical image segmentation

More Related Videos

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

42.7K
Author Spotlight: Advancing 3D Modeling for Enhanced Diagnosis and Treatment of Pulmonary Nodules in Early-Stage Lung Cancer
07:53

Author Spotlight: Advancing 3D Modeling for Enhanced Diagnosis and Treatment of Pulmonary Nodules in Early-Stage Lung Cancer

Published on: October 13, 2023

1.6K

Related Experiment Videos

Last Updated: Aug 23, 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

2.9K
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

42.7K
Author Spotlight: Advancing 3D Modeling for Enhanced Diagnosis and Treatment of Pulmonary Nodules in Early-Stage Lung Cancer
07:53

Author Spotlight: Advancing 3D Modeling for Enhanced Diagnosis and Treatment of Pulmonary Nodules in Early-Stage Lung Cancer

Published on: October 13, 2023

1.6K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Computer Vision

Background:

  • 3D medical image segmentation is vital for healthcare.
  • Existing 2D and 3D neural networks face an efficiency-accuracy trade-off.

Purpose of the Study:

  • To introduce a novel Mixture Convolutional Network (MixConvNet) for efficient and accurate 3D medical image segmentation.
  • To address the limitations of traditional convolutional blocks in 3D medical image segmentation.

Main Methods:

  • Proposed novel MixConv blocks that decompose 3D convolution into a mixture of 2D convolutions from different views.
  • MixConv blocks process volumetric data directly, learning intra-slice features while using fewer parameters and less computation than standard 3D convolutions.
  • Implemented a pre-training strategy with small patches followed by fine-tuning with large patches.

Main Results:

  • MixConvNet demonstrated superior performance compared to state-of-the-art methods like UNet3D, VNet, and nnUnet.
  • The proposed method achieved improved segmentation accuracy and efficiency.
  • Experiments were validated on the Decathlon Heart and Sliver07 datasets.

Conclusions:

  • MixConvNet offers a promising solution for 3D medical image segmentation, balancing efficiency and accuracy.
  • The novel MixConv block design effectively leverages 2D convolutions for improved 3D processing.
  • The findings suggest potential for enhanced clinical applications of automated medical image segmentation.