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

Assessing Impact of Data Quality in Early Post-Operative Glioblastoma Segmentation.

Journal of imaging·2026
Same author

Is tumor shape associated with molecular diagnosis, extent of resection, or postoperative focal deficits in diffuse low-grade gliomas?

Neuro-oncology advances·2025
Same author

Assessing Brain Injury After Extra-Axial Brain Tumor Resection: Circulating Neurofilament Light Levels, Neurological Deficits, and Fatigue.

Neurosurgery·2025
Same author

Disparities in end-of-life care and place of death in people with malignant brain tumors-A Swedish registry study.

Neuro-oncology practice·2025
Same author

Clinical application of machine-based deep learning in patients with radiologically presumed adult-type diffuse glioma grades 2 or 3.

Neuro-oncology advances·2024
Same author

The T1/T2 Ratio is Associated With Resectability in Patients With Isocitrate Dehydrogenase-Mutant Astrocytomas Central Nervous System World Health Organization Grades 2 and 3.

Neurosurgery·2024

Related Experiment Video

Updated: Sep 3, 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

A Feasibility Study on Deep Learning Based Brain Tumor Segmentation Using 2D Ellipse Box Areas.

Muhaddisa Barat Ali1, Xiaohan Bai1, Irene Yu-Hua Gu1

  • 1Department of Electrical Engineering, Chalmers University of Technology, 41296 Gothenburg, Sweden.

Sensors (Basel, Switzerland)
|July 27, 2022
PubMed
Summary

This study introduces a novel deep learning method for brain tumor segmentation using surrounding ellipse boxes instead of full annotations. This approach significantly reduces the need for expert annotation time while maintaining high segmentation accuracy.

Keywords:
2D ellipse box areasMR imagesbrain tumorsdeep learningglioma segmentationmulti-stream U-Net

More Related Videos

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
04:25

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

2.8K
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

8.6K

Related Experiment Videos

Last Updated: Sep 3, 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
Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
04:25

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

2.8K
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

8.6K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Neuro-oncology

Background:

  • Deep learning for brain tumor segmentation typically requires extensive manual annotation of tumor regions.
  • Accurate tumor annotation is time-consuming and requires specialized medical expertise.
  • Developing efficient segmentation methods is crucial for timely diagnosis and treatment planning.

Purpose of the Study:

  • To develop and evaluate a deep learning model for brain tumor segmentation using ellipse box annotations.
  • To assess the feasibility of training segmentation models with less precise, but more easily obtainable, bounding box data.
  • To compare the performance of this novel approach against traditional methods relying on fully annotated data.

Main Methods:

  • A multi-stream U-Net architecture was employed, leveraging multi-modal MRI data (T1, T1ce, T2, FLAIR).
  • The model was initially trained on unannotated MRIs using surrounding ellipse boxes (foreground/background).
  • A subsequent refinement phase utilized a small dataset (<20 patients) of fully annotated tumors for fine-tuning.

Main Results:

  • The proposed method achieved strong segmentation performance on glioma datasets (MICCAI BraTS’17 and US).
  • Dice scores of (0.8407, 0.9104) and accuracy of (83.88%, 88.47%) were obtained for the respective datasets.
  • Performance drop compared to fully annotated training was minimal (Dice: 0.0594, 0.0159; Accuracy: 8.78%, 2.61%).

Conclusions:

  • Training deep learning models for brain tumor segmentation using ellipse box annotations is a feasible alternative to full annotation.
  • This approach offers a practical trade-off between reduced annotation effort and a slight decrease in segmentation performance.
  • The method holds promise for improving the efficiency of AI-driven medical image analysis in neuro-oncology.