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

Iterated Residual Graph Convolutional Neural Network for Personalized Three-Dimensional Reconstruction of Left Myocardium from Cardiac MR Images.

Sensors (Basel, Switzerland)·2023
Same author

Two-Stage CNN Whole Heart Segmentation Combining Image Enhanced Attention Mechanism and Metric Classification.

Journal of digital imaging·2022
Same author

A Convolutional Neural Network Combining Discriminative Dictionary Learning and Sequence Tracking for Left Ventricular Detection.

Sensors (Basel, Switzerland)·2021
Same author

Myocardium Detection by Deep SSAE Feature and Within-Class Neighborhood Preserved Support Vector Classifier and Regressor.

Sensors (Basel, Switzerland)·2019
Same author

Automatic Vertebrae Localization and Identification by Combining Deep SSAE Contextual Features and Structured Regression Forest.

Journal of digital imaging·2019
Same author

Segmenting Brain Tissues from Chinese Visible Human Dataset by Deep-Learned Features with Stacked Autoencoder.

BioMed research international·2016

Related Experiment Video

Updated: Jul 25, 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

LcmUNet: A Lightweight Network Combining CNN and MLP for Real-Time Medical Image Segmentation.

Shuai Zhang1, Yanmin Niu1

  • 1School of Computer and Information Science, Chongqing Normal University, Chongqing 401331, China.

Bioengineering (Basel, Switzerland)
|June 28, 2023
PubMed
Summary

A new lightweight network, LcmUNet, offers fast and accurate medical image segmentation. This model uses convolutional and MLP layers to achieve real-time performance with fewer parameters, improving diagnostic capabilities.

Keywords:
CNNMLPUNetlightweight networkmedical image segmentation

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

444
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.5K

Related Experiment Videos

Last Updated: Jul 25, 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
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

444
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.5K

Area of Science:

  • Computer Vision
  • Medical Imaging
  • Artificial Intelligence

Background:

  • UNet and its variants are standard for medical image segmentation but suffer from high computational complexity.
  • Real-time medical image segmentation is crucial for timely diagnosis and therapy but is hindered by existing model limitations.

Purpose of the Study:

  • To introduce a lightweight medical image segmentation network (LcmUNet) that balances accuracy, parameter count, and computational efficiency.
  • To enable rapid, real-time medical image segmentation for clinical applications.

Main Methods:

  • Developed LcmUNet, a hybrid CNN-MLP architecture with convolutional layers (first three) and MLP layers (last two).
  • Incorporated a novel LDA module combining asymmetric, depth-wise separable convolutions, and attention mechanisms to reduce parameters while preserving feature extraction.
  • Introduced an LMLP module to enhance contextual and local information processing for improved accuracy and speed.

Main Results:

  • LcmUNet achieved real-time segmentation with only 1.49 million parameters, without pre-training.
  • Demonstrated strong performance across datasets: ISIC2018 (IoU 85.19%), BUSI (IoU 63.99%), and Kvasir-SEG (IoU 81.89%).
  • Achieved high precision and recall rates on all tested medical imaging datasets.

Conclusions:

  • LcmUNet provides an effective solution for real-time medical image segmentation.
  • The lightweight design and efficient modules make it suitable for resource-constrained environments and rapid clinical deployment.
  • The model shows significant potential for improving diagnostic workflows through accurate and fast segmentation.