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

Mechanisms of rectified gap junctional coupling enhancing pacemaking activity of biologically engineered pacemaker cells.

NPJ systems biology and applications·2026
Same author

Effect of Abnormal Calcium Dynamics on Heart Failure based on a Biophysical Modeling Study<sup></sup>.

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

Scribble-supervised method for cardiac tissue segmentation using position and temporal contrastive information.

Journal of medical imaging (Bellingham, Wash.)·2025
Same author

Advances in Structure Manipulation of Cobalt-Based Oxides for Catalytic Oxidation of Exhaust Gas.

Environmental science & technology·2025
Same author

Detection of Road Crack Images Based on Multistage Feature Fusion and a Texture Awareness Method.

Sensors (Basel, Switzerland)·2024
Same author

Ultrasound radiomics models based on multimodal imaging feature fusion of papillary thyroid carcinoma for predicting central lymph node metastasis.

Frontiers in oncology·2023

Related Experiment Video

Updated: May 23, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

348

SAMP-Net: a medical image segmentation network with split attention and multi-layer perceptron.

Xiaoxuan Ma1, Sihan Shan2, Dong Sui2

  • 1School of Electrical and Information Engineering, Beijing University of Civil Engineering and Architecture, No.15, Yongyuan Road, Huangcun Town, Daxing District, Beijing, 102616, China. maxiaoxuan@bucea.edu.cn.

Medical & Biological Engineering & Computing
|March 11, 2025
PubMed
Summary

A new medical image segmentation network enhances feature capture using attention and multilayer perceptrons (MLPs). This novel approach improves upon U-Net architectures for more efficient and accurate tumor segmentation in medical imaging.

Keywords:
Attention mechanismDepthwise separable convolutionsMedical image segmentationMultilayer perceptron (MLP)

More Related Videos

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

2.6K
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

2.3K

Related Experiment Videos

Last Updated: May 23, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

348
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

2.6K
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

2.3K

Area of Science:

  • Computer Vision
  • Medical Image Analysis
  • Deep Learning Architectures

Background:

  • Convolutional Neural Networks (CNNs) excel in computer vision, especially medical image segmentation.
  • U-Net architecture is a standard but has limitations in capturing multi-depth features due to uniform downsampling and simple convolutional layers.
  • These limitations hinder efficiency in rapid medical image processing.

Purpose of the Study:

  • To propose a novel segmentation network integrating attention mechanisms and multilayer perceptrons (MLPs).
  • To progressively capture and refine features at different network levels for improved medical image segmentation.
  • To enhance efficiency and performance compared to existing U-Net based models.

Main Methods:

  • Introduced Primary Feature Conservation (PFC) block in low-level layers to preserve spatial details during downsampling.
  • Incorporated Compact Attention Block (CAB) in mid-level layers for enhanced feature interaction via multi-path attention.
  • Utilized Shift MLP and Tokenized MLP blocks in high-level layers for improved local feature modeling and reduced computational complexity.

Main Results:

  • The proposed network demonstrated significant performance improvements on colorectal cancer tumor (CCI) and ISIC-2018 datasets.
  • Achieved average performance gains of 6.67% over U-Net, 5.53% over U-Net++, 10.18% over Swin-U-Net, 4.78% over Attention U-Net, and 3.55% over RA-U-Net.
  • The integration of attention and MLPs, along with specialized blocks, effectively addressed U-Net's limitations.

Conclusions:

  • The novel segmentation network offers superior performance and efficiency for medical image segmentation tasks.
  • The proposed architecture, incorporating PFC, CAB, Shift MLP, and Tokenized MLP blocks, represents a significant advancement over U-Net.
  • The findings suggest a promising direction for developing more effective deep learning models in medical image analysis.