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

Orthrus: toward evolutionary and functional RNA foundation models.

Nature methods·2026
Same author

Generalist biological artificial intelligence in modeling the language of life.

Nature biotechnology·2026
Same author

Language model-guided anticipation and discovery of mammalian metabolites.

Nature·2026
Same author

Harnessing artificial intelligence to advance CRISPR-based genome editing technologies.

Nature reviews. Genetics·2025
Same author

mRNABench: A curated benchmark for mature mRNA property and function prediction.

bioRxiv : the preprint server for biology·2025
Same author

Towards multimodal foundation models in molecular cell biology.

Nature·2025

Related Experiment Video

Updated: Jul 5, 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

2.8K

Segment anything in medical images.

Jun Ma1,2,3, Yuting He4, Feifei Li1

  • 1Peter Munk Cardiac Centre, University Health Network, Toronto, ON, Canada.

Nature Communications
|January 22, 2024
PubMed
Summary

MedSAM, a universal medical image segmentation foundation model, achieves superior accuracy and robustness across diverse tasks. Trained on over 1.5 million images, it enhances diagnostic tools and personalized treatment planning.

More Related Videos

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions
06:18

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

1.0K
A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment
10:39

A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment

Published on: May 24, 2022

2.4K

Related Experiment Videos

Last Updated: Jul 5, 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

2.8K
Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions
06:18

Author Spotlight: Segmentation and VR for Advanced Neurovascular Interventions

Published on: April 5, 2024

1.0K
A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment
10:39

A Label-Free Segmentation Approach for Intravital Imaging of Mammary Tumor Microenvironment

Published on: May 24, 2022

2.4K

Area of Science:

  • Medical imaging
  • Artificial intelligence
  • Computer vision

Background:

  • Medical image segmentation is crucial for clinical applications like diagnosis and treatment planning.
  • Current segmentation methods often lack generalizability across different imaging modalities and diseases.
  • A universal solution is needed to overcome the limitations of specialized models.

Purpose of the Study:

  • To introduce MedSAM, a foundation model for universal medical image segmentation.
  • To develop a model capable of generalizable segmentation across diverse medical imaging tasks.
  • To improve the accuracy and robustness of medical image analysis.

Main Methods:

  • Developed MedSAM, a foundation model for medical image segmentation.
  • Trained the model on a large-scale dataset of 1,570,263 image-mask pairs.
  • Dataset covered 10 imaging modalities and over 30 cancer types.

Main Results:

  • MedSAM demonstrated superior accuracy and robustness compared to modality-wise specialist models.
  • Evaluated on 86 internal and 60 external validation tasks.
  • Achieved accurate and efficient segmentation across a wide spectrum of medical imaging tasks.

Conclusions:

  • MedSAM offers a universal solution for medical image segmentation, addressing the lack of generalizability in existing methods.
  • The model has the potential to significantly advance diagnostic tools and personalized treatment planning.
  • MedSAM's performance highlights the power of foundation models in medical image analysis.