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

Coupling-matrix-designed dual-band sensor for high-accuracy water quantification in petroleum mixtures with self-organizing GMDH prediction.

Scientific reports·2026
Same author

An ultra-wide rejection suppressing cell for microwave applications.

PloS one·2026
Same author

Is ophthalmology a pain-free career? A systematic review and meta-analysis of musculoskeletal pain among ophthalmologists.

Canadian journal of ophthalmology. Journal canadien d'ophtalmologie·2026
Same author

Nano-Encapsulated Cinnamomum cassia Essential Oil in Combination With Fluconazole Exhibits Synergistic Activity Against Fluconazole-Resistant Nakaseomyces glabratus and Pichia kudriavzevii.

Veterinary medicine and science·2026
Same author

Thirty-Six-Month Multicentre Outcomes of Combined Phacoemulsification and Hydrus Microstent Implantation in Eyes With Normal Tension Glaucoma.

Journal of glaucoma·2026
Same author

High-sensitivity dual-band microstrip sensor for oil-water mixture analysis with RBF neural network optimization.

Scientific reports·2026

Related Experiment Video

Updated: May 22, 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

2.6K

Liver margin segmentation in abdominal CT images using U-Net and Detectron2: annotated dataset for deep learning

Mohammad Amir Sattari1, Seyed Abed Zonouri1, Ali Salimi2

  • 1Electrical Engineering Department, Faculty of Engineering, Razi University, Kermanshah, Iran.

Scientific Reports
|March 14, 2025
PubMed
Summary

Detectron2 significantly improved liver segmentation in CT scans, achieving a 0.974 Mask IoU, outperforming U-Net. This advancement aids in diagnosing liver diseases and developing automated medical imaging analysis.

Keywords:
Annotated datasetCT imagingDeep learningDetectron2Liver segmentationMedical diagnosticsU-Net

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

343
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.3K

Related Experiment Videos

Last Updated: May 22, 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

2.6K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

343
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.3K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Radiology

Background:

  • Accurate liver margin segmentation in computed tomography (CT) is crucial for diagnostics and treatment planning.
  • Complex liver anatomy and variability present significant segmentation challenges.
  • A new, large dataset of over 4,200 annotated abdominal CT images is now available.

Purpose of the Study:

  • To compare the performance of U-Net and Detectron2 deep neural network models for liver segmentation in CT images.
  • To evaluate the models' efficacy in handling complex anatomical variations.
  • To provide a validated dataset for enhancing and testing segmentation models.

Main Methods:

  • Utilized a dataset of 4,200+ expert-annotated abdominal CT images.
  • Implemented and compared two deep neural network models: U-Net and Detectron2.
  • Evaluated segmentation performance using the Mask Intersection over Union (Mask IoU) metric.

Main Results:

  • U-Net achieved a Mask IoU of 0.903, showing high efficacy in simpler cases.
  • Detectron2 outperformed U-Net, reaching a Mask IoU of 0.974.
  • Detectron2 demonstrated superior performance in segmenting complex cases with challenging liver boundary delineations.

Conclusions:

  • Detectron2 shows advanced potential for liver segmentation, especially in cases with significant anatomical variations.
  • The study provides a comparative analysis framework for deep learning models in medical image segmentation.
  • Findings support the development of automated systems for liver disease diagnosis and potential application to other organs.