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 Experiment Video

Updated: Jun 3, 2026

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

Cell segmentation in microscopy images using a SAM-based U-Net architecture and a novel dataset.

Md Shariful Alam1, Miriam Jackson2, Megan Lord2

  • 1School of Computer Science and Engineering, University of New South Wales, Sydney, Australia.

Computer Methods and Programs in Biomedicine
|June 1, 2026
PubMed
Summary

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

A lightweight network for segmenting tree-like structures in medical images.

Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society·2026
Same author

Multimodal animal health monitoring in extensive livestock production systems.

Frontiers in veterinary science·2026
Same author

OCT-based optic neuropathy diagnosis using explainable and privacy-preserving machine learning.

Scientific reports·2026
Same author

Medical hierarchical image classification via dual-geometry image-text learning.

Medical image analysis·2026
Same author

Multiple attention based deep multimodal fusion network for glaucoma and neurodegenerative disease diagnosis.

Scientific reports·2026
Same author

M2OTCA: Multiple-magnification optimal transport-based cross-attention learning for whole slide image classification.

Medical image analysis·2026

A new hybrid deep learning model, mSAMUNet, significantly improves cell instance segmentation in microscopy images. This advancement aids biomedical research by enhancing automated cell analysis for drug discovery and disease modeling.

Area of Science:

  • Biomedical image analysis
  • Computational biology
  • Deep learning for microscopy

Background:

  • Accurate cell instance segmentation is crucial for analyzing cell morphology, phenotypes, and dynamics in biomedical research.
  • Challenges include dense populations, blurred boundaries, and limited annotated datasets for deep learning models.

Purpose of the Study:

  • To introduce mCellSeg, a new dataset of expert-annotated microscopy images for cell instance segmentation.
  • To propose mSAMUNet, a novel hybrid neural network for improved cell instance segmentation.

Main Methods:

  • Developed mCellSeg dataset with 16,199 cells from HEK-293T and HUVEC lines.
  • Proposed mSAMUNet, integrating Segment Anything Model (SAM) with a multiscale U-Net encoder.
  • Benchmarked mSAMUNet against U-Net, StarDist, SAM, and microSAM using five-fold cross-validation.
Keywords:
Cell segmentationDeep learningMicroscopy images

More Related Videos

Automated Analysis of C. elegans Fluorescence Images using SegElegans
06:27

Automated Analysis of C. elegans Fluorescence Images using SegElegans

Published on: October 10, 2025

Related Experiment Videos

Last Updated: Jun 3, 2026

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

Automated Analysis of C. elegans Fluorescence Images using SegElegans
06:27

Automated Analysis of C. elegans Fluorescence Images using SegElegans

Published on: October 10, 2025

Main Results:

  • mSAMUNet outperformed existing methods across multiple datasets and metrics (F1 score, SA50, SA75, mean SA).
  • Achieved superior performance on the mCellSeg dataset (F1 = 0.7071 vs. microSAM's 0.6994) and NeurIPS22 dataset (F1 = 0.8675 vs. microSAM's 0.8483).
  • Demonstrated the effectiveness of combining CNNs and Transformers for biomedical image segmentation.

Conclusions:

  • mSAMUNet offers a robust and accurate solution for microscopy cell instance segmentation, exceeding state-of-the-art performance.
  • The mCellSeg dataset and mSAMUNet model provide valuable resources for automated cell analysis in drug discovery and disease modeling.