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

Migration Intent Among Early-Career Doctors in Pakistan: A Cross-Sectional Study of Influencing Factors at a Tertiary Care Hospital.

Cureus·2026

Same author

Compact ferroelectric domain-ensemble model linking PUND switching, wake-up, and plasmonic optical readout.

Optics express·2026

Same author

When Light Drives Discovery: Plasmonics at the Frontier of Energy Materials.

Chemistry of materials : a publication of the American Chemical Society·2026

Same author

HfO<sub>2</sub>-based memristive synapses with asymmetrically extended p-n heterointerfaces for highly energy-efficient neuromorphic hardware.

Science advances·2026

Same author

Resetting the Drift of Oxygen Vacancies in Ultrathin HZO Ferroelectric Memories by Electrical Pulse Engineering.

Small science·2025

Same author

Scanning Plasmon-Enhanced Microscopy for Simultaneous Optoelectrical Characterization.

ACS nano·2024

Search research articles

Related Experiment Video

Updated: Apr 2, 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

3D-ViT-UNet: 3D Vision transformer based Unet-like model for Volumetric Brain Tumor Segmentation.

Sikandar Afridi¹, Atif Jan¹, Muhammad Abeer Irfan²

¹Department of Electrical Engineering, University of Engineering and Technology, Peshawar, Pakistan.

PLOS Digital Health

|March 31, 2026

Summary

A new 3D Vision Transformer U-Net (3D-ViT-UNet) improves 3D brain tumor segmentation. This model enhances accuracy and efficiency for clinical diagnosis and therapy planning.

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Multicolor 3D Printing of Complex Intracranial Tumors in Neurosurgery

Multicolor 3D Printing of Complex Intracranial Tumors in Neurosurgery

Published on: January 11, 2020

Related Experiment Videos

Last Updated: Apr 2, 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

Multicolor 3D Printing of Complex Intracranial Tumors in Neurosurgery

Multicolor 3D Printing of Complex Intracranial Tumors in Neurosurgery

Published on: January 11, 2020

Area of Science:

Medical Imaging and Artificial Intelligence
Neuro-oncology and Computational Pathology

Background:

Accurate 3D medical image segmentation is vital for brain tumor diagnosis and treatment planning.
Traditional Convolutional Neural Networks (CNNs) struggle with global context and long-range dependencies in volumetric data.
Transformer models offer a promising alternative for capturing spatial dependencies in medical imaging.

Purpose of the Study:

To introduce 3D-ViT-UNet, a novel U-shaped Vision Transformer (ViT)-based architecture for end-to-end volumetric brain tumor segmentation.
To address the limitations of CNNs in capturing global contextual information for improved segmentation accuracy.
To develop an efficient and effective model for clinical applications in brain tumor delineation.

Main Methods:

Proposed a novel U-shaped encoder-decoder architecture, 3D-ViT-UNet, leveraging Vision Transformers (ViT).
Incorporated 3D Window Multi-Head Self-Attention (3D-W-MSA) for local feature extraction and 3D Dilated-Window Multi-Head Self-Attention (3D-DW-MSA) for global feature extraction, reducing computational cost.
Integrated a dynamic position encoding strategy to preserve positional information and mitigate transformer limitations.

Main Results:

Achieved state-of-the-art (SOTA) performance on the BraTS 2020 dataset for brain tumor segmentation.
Obtained a superior average Dice Similarity Coefficient (DSC) of 84.81% and Hausdorff Distance (HD) of 4.87 mm.
Demonstrated reduced computational complexity (FLOPs) and a smaller model size compared to existing methods, with improved boundary delineation.

Conclusions:

3D-ViT-UNet effectively captures both local and global features for accurate volumetric brain tumor segmentation.
The model offers high accuracy, efficiency, and improved delineation capabilities, making it suitable for clinical applications.
The proposed architecture represents a significant advancement in automated brain tumor segmentation using deep learning.