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

Effect of an annotation-free artificial intelligence system for simultaneous detection and diagnosis on breast ultrasonography: a multireader, multicase study across diverse healthcare professionals.

Ultrasonography (Seoul, Korea)·2026
Same author

Prediction of Anthracycline-induced Cardiotoxicity Using Cardiac MRI Parameters: An Animal Study.

Radiology. Cardiothoracic imaging·2026
Same author

A comprehensive clinical, molecular, radiological, and surgical analysis for predicting prognosis in IDH-mutant astrocytoma patients.

Neuro-oncology·2026
Same author

An AI-driven, wearable, conformal ring system for real-time and user-independent sign language interpretation.

Science advances·2026
Same author

Evaluation of breast lesions using a newly developed 3D breast ultrasound imaging device: a preliminary study on efficacy and validity.

Scientific reports·2026
Same author

Higher cerebral metabolic rate of oxygen derived from dynamic susceptibility contrast MRI is an independent predictor of poor survival in IDH-wildtype glioblastoma.

Neuroradiology·2026

Related Experiment Video

Updated: Aug 2, 2025

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

2.5K

Intelligent noninvasive meningioma grading with a fully automatic segmentation using interpretable multiparametric

Yohan Jun1,2, Yae Won Park3, Hyungseob Shin4

  • 1Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, USA.

European Radiology
|April 13, 2023
PubMed
Summary
This summary is machine-generated.

A new deep learning (DL) model accurately grades meningiomas using MRI scans, outperforming human readers. This interpretable model also segments tumors, offering a robust, noninvasive diagnostic tool.

Keywords:
Deep learningInterpretableMagnetic resonance imagingMeningioma, grading

More Related Videos

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

2.8K
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
06:48

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Published on: January 7, 2019

9.0K

Related Experiment Videos

Last Updated: Aug 2, 2025

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

2.5K
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

2.8K
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
06:48

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Published on: January 7, 2019

9.0K

Area of Science:

  • Neuroimaging
  • Artificial Intelligence in Medicine
  • Oncology

Background:

  • Meningiomas are the most common primary central nervous system tumors.
  • Accurate grading of meningiomas is crucial for treatment planning and prognosis.
  • Current grading methods can be invasive or lack precision.

Purpose of the Study:

  • To develop and validate an interpretable multiparametric deep learning (DL) model for automatic, noninvasive grading and segmentation of meningiomas.
  • To assess the diagnostic performance of the DL model compared to human readers.

Main Methods:

  • A two-stage DL model using three-dimensional U-net and ResNet was trained on multiparametric MRI (T1C and T2) from 257 patients.
  • The model was validated on an external dataset of 61 patients.
  • Relevance-weighted Class Activation Mapping (RCAM) was employed for model interpretability.

Main Results:

  • The combined T1C and T2 DL model achieved a Dice coefficient of 0.910 for segmentation on external validation.
  • The model demonstrated superior grading performance with an AUC of 0.770 and 72.1% accuracy, outperforming human readers (AUCs 0.675-0.690).
  • RCAM analysis revealed the model focused on tumor margin features for grading.

Conclusions:

  • An interpretable multiparametric DL model combining T1C and T2 MRI data enables fully automatic meningioma grading and segmentation.
  • The DL model exhibits robust performance and surpasses human reader capabilities.
  • The model's interpretability through RCAM highlights its ability to identify clinically relevant features at the tumor margin.