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

A Progressive Gaussian Splatting Framework for Monocular-Only High-Fidelity Surgical Scene Reconstruction.

IEEE transactions on bio-medical engineering·2026
Same author

BCIRT: Backscattering-corrected implicit representation tomography.

Medical image analysis·2026
Same author

Quasi-static Elastography-driven Automated Robotic Ultrasound Screening and Localization.

IEEE transactions on bio-medical engineering·2025
Same author

SurGSplat: Progressive Geometry-Constrained Gaussian Splatting for Surgical Scene Reconstruction.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
Same author

Toward Clinical Applications of Intelligent Robotic Ultrasound Systems.

IEEE reviews in biomedical engineering·2025
Same author

Amber Light-Assisted CBI Endoscopy for Superior Deep Vascular Visualization and Blood-Containing Tissue Depth Differentiation.

IEEE transactions on bio-medical engineering·2025

Related Experiment Video

Updated: Jan 9, 2026

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning
09:31

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning

Published on: April 28, 2022

3.5K

UltraLight VM-UNet: Parallel Vision Mamba significantly reduces parameters for skin lesion segmentation.

Renkai Wu1,2,3, Yinghao Liu4, Guochen Ning5

  • 1Department of Geriatrics, Medical Center on Aging of Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai 200025, China.

Patterns (New York, N.Y.)
|December 2, 2025
PubMed
Summary

This study introduces UltraLight Vision Mamba UNet (UltraLight VM-UNet), a computationally efficient model for medical image segmentation. It achieves competitive performance with minimal parameters, ideal for mobile medical devices.

Keywords:
Mambalightweight modelskin lesion segmentation

More Related Videos

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

2.7K
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

3.3K

Related Experiment Videos

Last Updated: Jan 9, 2026

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning
09:31

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning

Published on: April 28, 2022

3.5K
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

2.7K
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

3.3K

Area of Science:

  • Medical Imaging
  • Computer Vision
  • Artificial Intelligence

Background:

  • Traditional segmentation models often require complex modules, limiting their use in resource-constrained environments like mobile medical devices.
  • State-space models, such as Mamba, are emerging as efficient alternatives to traditional CNNs and Transformers.

Purpose of the Study:

  • To explore parameter influence in Mamba for medical image segmentation.
  • To develop a computationally efficient segmentation model suitable for mobile medical applications.

Main Methods:

  • Proposed the UltraLight Vision Mamba UNet (UltraLight VM-UNet) architecture.
  • Introduced a Parallel Vision Mamba (PVM) Layer for efficient feature processing.
  • Evaluated the model on three public skin lesion segmentation datasets.

Main Results:

  • UltraLight VM-UNet achieved competitive segmentation performance.
  • The model demonstrated exceptionally low computational complexity with only 0.049M parameters and 0.060 GFLOPs.
  • The PVM Layer maintained performance while reducing computational load.

Conclusions:

  • UltraLight VM-UNet offers a highly efficient solution for medical image segmentation.
  • The proposed PVM Layer effectively reduces computational complexity without sacrificing performance.
  • This model is well-suited for deployment on mobile medical devices with limited computational resources.