Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Mar 12, 2026

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Class-balanced dermoscopic lesion segmentation using MoG-LISA and optimized Swin-UNet via the GM-FDE framework.

S Muthamil Selvan¹, R Kavitha²

¹Department of Computer Science and Engineering, Faculty of Engineering and Technology, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India.

|March 11, 2026

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

Competitive roles of plasticity, elasticity, and radiation in heat/mass transfer of Casson-Williamson-Maxwell nanofluid over a porous stretching sheet.

Discover nano·2026

Same author

Effect of Aluminum Oxide Air Abrasion on Shear Bond Strength of Composite Resin to Human Dentin Using Two Adhesive Systems.

Journal of pharmacy & bioallied sciences·2026

Same author

Comparative Evaluation of Apically Extruded Debris Using Two Rotary and Two Reciprocating File Systems: An <i>In Vitro</i> Study.

Journal of pharmacy & bioallied sciences·2025

Same author

Soret Effect and Chemical Process on MHD Oscillatory Flow in a Physiological Fluid.

Applied bionics and biomechanics·2025

Same author

Enhancing e-waste management: a novel light gradient AdaBoost support vector classification approach.

Environmental monitoring and assessment·2025

Same author

Dynamic feature selection and quantum representation for precise heart disease prediction: Quantum-HeartDiseaseNet approach.

Computer methods in biomechanics and biomedical engineering·2025

This study introduces a new framework for skin lesion segmentation, improving early melanoma diagnosis. The approach enhances segmentation accuracy and boundary detection for better dermatological image analysis.

Area of Science:

Dermatological image analysis
Medical image processing
Deep learning for medical imaging

Background:

Automatic skin lesion segmentation is crucial for early melanoma diagnosis and treatment planning.
Current deep learning methods struggle with class imbalance, morphological variability, and poor boundary delineation.
Addressing these challenges is vital for advancing dermatological image analysis.

Purpose of the Study:

To propose a unified framework combining data augmentation and model optimization for improved skin lesion segmentation.
To enhance the handling of class imbalance and morphological variability in skin lesion datasets.
To achieve precise boundary detection in dermatological images.

Main Methods:

Developed MoG-LISA (morphology-guided latent interpolation and synthesis for lesion augmentation) to generate high-fidelity synthetic samples, enriching underrepresented classes.

Keywords:

Dermatology Health sciences Health technology Medicine

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Related Experiment Videos

Last Updated: Mar 12, 2026

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Implemented CB-SwinGMO (Class-Balanced Swin-UNet optimization using geometric mean-driven feedback evolutionary framework) for multi-objective evolutionary optimization of Swin-UNet parameters.

Focused on improving generalization and boundary detection capabilities of the segmentation model.

Main Results:

The proposed framework achieved a Dice Similarity Coefficient (DSC) of 93.8% and IoU of 91.2% on the SIIM-ISIC dataset.
Boundary accuracy reached 92.7%, with a Hausdorff distance reduction of up to 4.3 pixels.
Demonstrated superior performance in segmenting skin lesions, particularly for underrepresented classes.

Conclusions:

The unified framework effectively addresses challenges in skin lesion segmentation, including class imbalance and boundary delineation.
MoG-LISA and CB-SwinGMO significantly improve the accuracy and precision of dermatological image analysis.
The approach shows strong potential for enhancing early melanoma diagnosis and treatment planning.