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

Artificial neural network potential for Au<sub>20</sub>clusters based on the first-principles.

Journal of physics. Condensed matter : an Institute of Physics journal·2022
Same author

Odor-producing response pattern by four typical freshwater algae under stress: Acute microplastic exposure as an example.

The Science of the total environment·2022
Same author

Multi-Omics Analysis of the Gut-Liver Axis Reveals the Mechanism of Liver Injury in Colitis Mice.

Frontiers in immunology·2022
Same author

'C-type' closed state and gating mechanisms of K2P channels revealed by conformational changes of the TREK-1 channel.

Journal of molecular cell biology·2022
Same author

Treatment of Radiation-Induced Brain Necrosis.

Oxidative medicine and cellular longevity·2022
Same author

Effect of Nano-Silver on Formation of Marine Snow and the Underlying Microbial Mechanism.

Environmental science & technology·2022

Related Experiment Video

Updated: Aug 29, 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 deep learning-based self-adapting ensemble method for segmentation in gynecological brachytherapy.

Zhen Li1, Qingyuan Zhu1, Lihua Zhang1

  • 1Shanghai Jiao Tong University Affiliated Sixth People's Hospital, Xuhui District, Shanghai, China.

Radiation Oncology (London, England)
|September 5, 2022
PubMed
Summary
This summary is machine-generated.

nnU-Net achieved superior auto-segmentation of organs at risk (OARs) and high-risk clinical tumor volumes (HRCTVs) in gynecological brachytherapy. The 3D-Cascade U-Net model demonstrated the highest accuracy, improving treatment planning efficiency.

Keywords:
Auto-segmentationDeep learningGynecological cancerHigh-dose-rate brachytherapy

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

485
Stereotactic Radiosurgery for Gynecologic Cancer
10:35

Stereotactic Radiosurgery for Gynecologic Cancer

Published on: April 17, 2012

18.3K

Related Experiment Videos

Last Updated: Aug 29, 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
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

485
Stereotactic Radiosurgery for Gynecologic Cancer
10:35

Stereotactic Radiosurgery for Gynecologic Cancer

Published on: April 17, 2012

18.3K

Area of Science:

  • Medical Imaging
  • Radiotherapy Oncology
  • Artificial Intelligence in Medicine

Background:

  • Accurate segmentation of organs at risk (OARs) and high-risk clinical tumor volumes (HRCTVs) is critical for effective gynecological brachytherapy.
  • The manual contouring process is time-intensive, posing challenges for online treatment planning and dose escalation due to steep dose gradients.

Purpose of the Study:

  • To apply a self-configured ensemble deep learning method (nnU-Net) for fast and reproducible auto-segmentation of OARs and HRCTVs in gynecological cancer.
  • To evaluate the accuracy and efficiency of nnU-Net compared to previous methods and manual contouring.

Main Methods:

  • nnU-Net, a deep convolutional neural network, was employed to segment the bladder, rectum, and HRCTV on CT images.
  • Three nnU-Net architectures (2D U-Net, 3D U-Net, 3D-Cascade U-Net) were trained and ensembled, with 207 cases for training and 30 for testing.
  • Segmentation performance was quantitatively assessed using Dice Similarity Coefficient (DSC), 95% Hausdorff Distance (HD95%), and Average Surface Distance (ASD), alongside qualitative and dosimetric evaluations.

Main Results:

  • nnU-Net demonstrated high segmentation accuracy, outperforming previously reported methods for bladder and rectum.
  • The 3D-Cascade U-Net achieved the best performance: Bladder (DSC: 0.936, HD95%: 3.503), Rectum (DSC: 0.831, HD95%: 7.579), and HRCTV (DSC: 0.836, HD95%: 7.42).
  • Qualitative analysis revealed that over 76% of segmentations had no or minor visually detectable errors.

Conclusions:

  • nnU-Net shows superiority for segmenting OARs and HRCTVs in gynecological brachytherapy.
  • The 3D-Cascade U-Net architecture within the nnU-Net framework offers the highest segmentation accuracy across diverse patient anatomies and applicators.
  • This automated approach has the potential to significantly improve the efficiency and reproducibility of treatment planning.