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

Author Correction: A novel 2-phase residual U-net algorithm combined with optimal mass transportation for 3D brain tumor detection and segmentation.

Scientific reports·2022
Same author

A novel 2-phase residual U-net algorithm combined with optimal mass transportation for 3D brain tumor detection and segmentation.

Scientific reports·2022
Same author

Use of enhanced T2 star-weighted angiography (ESWAN) and R2* values to distinguish ovarian cysts due to endometriosis from other causes.

Abdominal imaging·2014
Same author

Fluorescence and energy transfer properties of heterometallic lanthanide-titanium oxo clusters coordinated with anthracenecarboxylate ligands.

Dalton transactions (Cambridge, England : 2003)·2014
Same author

Chronic intermittent hypobaric hypoxia ameliorates endoplasmic reticulum stress mediated liver damage induced by fructose in rats.

Life sciences·2014
Same author

Magnetic nanoparticle clusters for photothermal therapy with near-infrared irradiation.

Biomaterials·2014

Related Experiment Video

Updated: Oct 25, 2025

Patient-Specific Polyvinyl Alcohol Phantom Fabrication with Ultrasound and X-Ray Contrast for Brain Tumor Surgery Planning
08:41

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

Published on: July 14, 2020

8.7K

3D brain tumor segmentation using a two-stage optimal mass transport algorithm.

Wen-Wei Lin1, Cheng Juang2, Mei-Heng Yueh3

  • 1Department of Applied Mathematics, National Yang Ming Chiao Tung University, Hsinchu, 300, Taiwan.

Scientific Reports
|August 11, 2021
PubMed
Summary

Optimal mass transport preprocesses brain tumor data, transforming irregular shapes into cubes for U-Net analysis. This method significantly reduces data size while preserving detail, improving tumor segmentation accuracy.

More Related Videos

Live Imaging of Microtubule Dynamics in Glioblastoma Cells Invading the Zebrafish Brain
09:29

Live Imaging of Microtubule Dynamics in Glioblastoma Cells Invading the Zebrafish Brain

Published on: July 29, 2022

2.9K
Multicolor 3D Printing of Complex Intracranial Tumors in Neurosurgery
14:15

Multicolor 3D Printing of Complex Intracranial Tumors in Neurosurgery

Published on: January 11, 2020

7.3K

Related Experiment Videos

Last Updated: Oct 25, 2025

Patient-Specific Polyvinyl Alcohol Phantom Fabrication with Ultrasound and X-Ray Contrast for Brain Tumor Surgery Planning
08:41

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

Published on: July 14, 2020

8.7K
Live Imaging of Microtubule Dynamics in Glioblastoma Cells Invading the Zebrafish Brain
09:29

Live Imaging of Microtubule Dynamics in Glioblastoma Cells Invading the Zebrafish Brain

Published on: July 29, 2022

2.9K
Multicolor 3D Printing of Complex Intracranial Tumors in Neurosurgery
14:15

Multicolor 3D Printing of Complex Intracranial Tumors in Neurosurgery

Published on: January 11, 2020

7.3K

Area of Science:

  • Medical image processing
  • Computational anatomy
  • Machine learning in oncology

Background:

  • Brain tumor segmentation is crucial for diagnosis and treatment planning.
  • Preprocessing irregular 3D brain data for deep learning models presents significant challenges.
  • Existing methods often struggle with preserving anatomical details during transformation.

Purpose of the Study:

  • To introduce and evaluate a novel preprocessing technique for brain tumor datasets using Optimal Mass Transport (OMT).
  • To adapt irregular 3D brain tumor data into a standardized cubic format suitable for 3D Convolutional Neural Networks (CNNs).
  • To assess the impact of this preprocessing on the accuracy and efficiency of brain tumor segmentation.

Main Methods:

  • Application of a two-stage Optimal Mass Transport (TSOMT) procedure to transform 3D brain tumor datasets into a cubic format.
  • Training of three U-Net models for segmenting Whole Tumor (WT), Tumor Core (TC), and Enhancing Tumor (ET) on the transformed cubic data.
  • Utilizing an inverse TSOMT method to map segmented results back to the original brain space.
  • Incorporating postprocessing steps, including rotation, to further enhance segmentation accuracy during testing.

Main Results:

  • Successful transformation of original brain tumor datasets into a cubic format, achieving a 76.6% reduction in voxel count with minimal distortion.
  • High training accuracy achieved (0.9822 for WT) and excellent Dice scores during training (0.9781 for WT, 0.9637 for TC, 0.9305 for ET).
  • Conversion loss between forward and inverse TSOMT methods was less than one percent, demonstrating the method's reversibility.
  • Testing accuracy improved by 1-2% with postprocessing, and the entire segmentation process completed in 200 seconds per dataset.

Conclusions:

  • TSOMT is a robust and effective method for preprocessing irregular 3D brain tumor data for deep learning applications.
  • This approach significantly improves segmentation accuracy and efficiency compared to traditional methods.
  • The minimal conversion loss and high segmentation scores validate the clinical potential of TSOMT in neuro-oncology.