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: Jan 17, 2026

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

FocusGate-Net: A dual-attention guided MLP-convolution hybrid network for accurate and efficient medical image

Şafak Kılıç1,2

  • 1School of Computer Science, CHART Laboratory, University of Nottingham, Nottingham, United Kingdom.

Plos One
|September 25, 2025
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

Multi-scale feature integration with enhanced cytomorph for high-accuracy cervical cytology classification.

PloS one·2026
Same author

A pruned and parameter-efficient Xception framework for skin cancer classification.

PloS one·2026
Same author

DiagNeXt: A Two-Stage Attention-Guided ConvNeXt Framework for Kidney Pathology Segmentation and Classification.

Journal of imaging·2025
Same author

Enhanced Vertebra-GNN: A Graph-Based Deep Learning Framework for Anatomically Informed Spinal Disease Classification.

Journal of imaging informatics in medicine·2025
Same author

Deep feature engineering for accurate sperm morphology classification using CBAM-enhanced ResNet50.

PloS one·2025
Same author

Attention-Based Dual-Path Deep Learning for Blood Cell Image Classification Using ConvNeXt and Swin Transformer.

Journal of imaging informatics in medicine·2025

FocusGate-Net, a novel hybrid architecture, enhances medical image segmentation by combining MLP efficiency with convolutional and attention mechanisms. This approach achieves high accuracy and speed, suitable for real-time clinical applications.

Area of Science:

  • Computer Vision
  • Medical Imaging Analysis
  • Artificial Intelligence

Background:

  • Current Convolutional Neural Networks (CNNs) and Transformers face challenges in balancing medical image segmentation accuracy, inference speed, and architectural complexity.
  • Lightweight Multi-Layer Perceptron (MLP)-based methods offer efficiency but often struggle with fine-grained spatial context and boundary localization.
  • There is a need for hybrid architectures that combine MLP computational efficiency with the spatial feature extraction capabilities of other modules for improved medical image segmentation.

Purpose of the Study:

  • To introduce FocusGate-Net, a novel hybrid architecture for robust medical image segmentation.
  • To leverage the strengths of MLPs for spatial dependency modeling and enhance feature selectivity using attention mechanisms.
  • To achieve high segmentation accuracy with low latency and minimal architectural complexity for real-time clinical deployment.

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

735

Related Experiment Videos

Last Updated: Jan 17, 2026

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

735

Main Methods:

  • Developed FocusGate-Net, a hybrid architecture integrating shifted token MLP blocks, convolutional feature extractors, and dual-attention mechanisms (Convolutional Block Attention Module - CBAM and Attention Gate - AG).
  • Evaluated the model on three diverse medical image datasets: ISIC2018 (skin lesions), PH2 (dermatoscopic images), and Kvasir-SEG (polyps).
  • Conducted comprehensive ablation studies to validate the contribution of each architectural component.

Main Results:

  • FocusGate-Net achieved superior performance compared to state-of-the-art models (UNet, UNet++, ResUNet) on the ISIC2018 dataset, with a Dice coefficient of 92.47% and IoU of 86.36%.
  • Demonstrated exceptional cross-dataset generalization capabilities, achieving Dice scores of 97.25% on PH2 and 94.83% on Kvasir-SEG.
  • Ablation studies confirmed the effectiveness of the hybrid design and attention mechanisms.

Conclusions:

  • FocusGate-Net effectively balances segmentation accuracy, inference speed, and architectural simplicity for medical image segmentation.
  • The proposed MLP-based hybrid architecture with attention mechanisms shows significant potential for improving accuracy and efficiency in clinical settings.
  • The model's strong performance and generalization capabilities highlight its suitability for real-time medical image analysis applications.