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

Tumor-Targeted pan-RAS Inhibition as a Novel Biologic Therapy for Diffuse Midline Glioma.

Neuro-oncology·2026

Same author

Tumor-derived cell-free DNA detected in cerebrospinal fluid enables minimally invasive profiling of pediatric brain tumors.

The Journal of clinical investigation·2026

Same author

Genetic and Congenital Cytomegalovirus-Related Hearing Loss in Children: Volumetric MRI Analysis of Auditory and Visual Cortices.

AJNR. American journal of neuroradiology·2026

Same author

Ultrasound-Enhanced Drug Delivery in Pediatric Neuro-Oncology: A New Therapeutic Strategy.

Pharmaceutics·2026

Same author

Tempered Optimism: Advances in the Precision Medicine Era for Pediatric Low-Grade Glioma.

Neuro-oncology·2026

Same author

TRAM-01: A phase 2 study of trametinib for pediatric patients with neurofibromatosis type 1 and plexiform neurofibromas.

Neuro-oncology·2026

Search research articles

Related Experiment Video

Updated: Jun 4, 2025

Patient-Specific Polyvinyl Alcohol Phantom Fabrication with Ultrasound and X-Ray Contrast for Brain Tumor Surgery Planning

Patient-Specific Polyvinyl Alcohol Phantom Fabrication with Ultrasound and X-Ray Contrast for Brain Tumor Surgery Planning

Published on: July 14, 2020

Generating 3D brain tumor regions in MRI using vector-quantization Generative Adversarial Networks.

Meng Zhou¹, Matthias W Wagner², Uri Tabori³

¹Department of Computer Science, University of Toronto, 40 St George St., Toronto, M5S 2E4, ON, Canada; Neurosciences & Mental Health Research Program, The Hospital for Sick Children, 686 Bay St., Toronto, M5G 0A4, ON, Canada.

Computers in Biology and Medicine

|December 19, 2024

Summary

This summary is machine-generated.

This study introduces a new deep learning framework using Generative Adversarial Networks (GANs) to create 3D brain tumor regions of interest (ROIs) for imbalanced MRI datasets. The method enhances tumor classification accuracy, aiding in the diagnosis of rare brain tumors.

Keywords:

Brain MRI Deep learning Generative Adversarial Networks Image generation Pediatric low grade glioma Transformer

More Related Videos

Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging

Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging

Published on: June 7, 2020

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: Jun 4, 2025

Patient-Specific Polyvinyl Alcohol Phantom Fabrication with Ultrasound and X-Ray Contrast for Brain Tumor Surgery Planning

Patient-Specific Polyvinyl Alcohol Phantom Fabrication with Ultrasound and X-Ray Contrast for Brain Tumor Surgery Planning

Published on: July 14, 2020

Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging

Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging

Published on: June 7, 2020

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 image analysis
Deep learning
Artificial intelligence in healthcare

Background:

Deep learning, especially Generative Adversarial Networks (GANs), enhances medical image analysis by generating realistic training data.
Current GAN approaches often generate entire image volumes, which is less effective for brain tumor classification than focusing on the region of interest (ROI).
Classifying brain tumor ROIs using deep learning on MRI is more effective than classifying entire volumes.

Purpose of the Study:

To develop a novel framework for generating high-resolution, diverse 3D brain tumor ROIs using vector-quantization GAN and a transformer model.
To augment imbalanced datasets, specifically low-grade glioma (LGG) and pediatric LGG (pLGG) tumor ROIs, for improved classification.
To address the challenge of imbalanced data in medical image analysis for brain tumor diagnosis.

Main Methods:

Utilized a novel framework combining vector-quantization GAN and a transformer with masked token modeling.
Generated 3D brain tumor ROIs for two imbalanced datasets: BraTS 2019 (LGG) and an internal pediatric LGG (pLGG) dataset.
Applied the generated ROIs to augment minority classes in the datasets for tumor ROI classification.

Main Results:

The proposed method demonstrated superior qualitative and quantitative performance compared to baseline models.
Achieved a 6.4% increase in Area Under the ROC Curve (AUC) on the BraTS 2019 dataset and a 4.3% increase in AUC on the pLGG dataset.
Effectively addressed data imbalance issues, improving brain tumor classification accuracy.

Conclusions:

The generated 3D brain tumor ROIs are effective in mitigating imbalanced data problems in medical image analysis.
The novel framework shows significant potential for improving the accuracy of rare brain tumor diagnosis using MRI scans.
This approach facilitates more accurate classification of brain tumors by providing augmented, high-quality training data.