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

ScopeViewer: A Browser-Based Solution for Visualizing Large Biological Images.

GigaScience·2026
Same author

MicNet: integrating spatially resolved transcriptomes and pathology images by contrastive deep neural network.

Genome biology·2026
Same author

Oligonucleotide genetics for <i>Pseudomonas aeruginosa</i> enables high throughput hypomorph screening.

bioRxiv : the preprint server for biology·2026
Same author

Computational identification of migrating T cells in spatial transcriptomics data.

JCI insight·2026
Same author

BiGER: Bayesian rank aggregation in genomics with extended ranking schemes.

Nature communications·2026
Same author

A microbiota-derived bile acid overcomes antibiotic-induced hyporesponsiveness to immune checkpoint therapy by enhancing CD8 <sup>+</sup> T cell antitumor immunity.

bioRxiv : the preprint server for biology·2026

Related Experiment Video

Updated: Jun 21, 2025

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

8.9K

Enhancing Medical Imaging Segmentation with GB-SAM: A Novel Approach to Tissue Segmentation Using Granular Box

Ismael Villanueva-Miranda1, Ruichen Rong1, Peiran Quan1

  • 1Quantitative Biomedical Research Center, Peter O'Donnell Jr. School of Public Health, University of Texas Southwestern Medical Center, Dallas, TX 75390, USA.

Cancers
|July 13, 2024
PubMed
Summary
This summary is machine-generated.

The Granular Box Prompt Segment Anything Model (GB-SAM) improves digital pathology image analysis by accurately segmenting glands with less training data. This foundation model enhances efficiency and reduces reliance on expert annotations for medical imaging tasks.

Keywords:
digital pathologyfoundation modelshistopathologypathology imagesegmentation

More Related Videos

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

24.5K
Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench
11:38

Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench

Published on: August 23, 2017

9.8K

Related Experiment Videos

Last Updated: Jun 21, 2025

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

8.9K
From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

24.5K
Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench
11:38

Volume Segmentation and Analysis of Biological Materials Using SuRVoS Super-region Volume Segmentation Workbench

Published on: August 23, 2017

9.8K

Area of Science:

  • Digital Pathology
  • Medical Imaging Analysis
  • Foundation Models

Background:

  • Traditional digital pathology methods require extensive manual annotations, hindering model development.
  • Foundation models offer generalization with few-shot learning, addressing adaptation challenges in medical imaging.
  • Automated annotation is crucial for efficiency but often struggles with complex morphological regions.

Purpose of the Study:

  • To introduce the Granular Box Prompt Segment Anything Model (GB-SAM), an advancement over SAM.
  • To enhance automated annotation efficiency by reducing dependency on expert pathologists.
  • To improve segmentation accuracy for individual glands in digital pathology using limited training data.

Main Methods:

  • Fine-tuned the Segment Anything Model (SAM) using granular box prompts derived from ground truth masks.
  • Utilized small box regions for localized analysis, replacing large bounding boxes on H&E-stained image patches.
  • Compared GB-SAM performance against U-Net across CRAG, GlaS, and Camelyon16 histopathological datasets.

Main Results:

  • GB-SAM outperformed U-Net on the CRAG dataset (Dice 0.885 vs. 0.857 with 25% data).
  • Demonstrated superior generalization on Camelyon16 lymph node segmentation (Dice 0.740 vs. U-Net's 0.491).
  • Showed competitive performance against SAM-Path and Med-SAM, particularly with limited training data.

Conclusions:

  • GB-SAM offers advanced segmentation capabilities in digital pathology with reduced data dependency.
  • The granular box prompt method enhances localized analysis and segmentation accuracy.
  • GB-SAM shows significant potential for practical deployment in resource-limited settings.