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

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

Genome-wide identification and expression profiling of the HMGR gene family in Bupleurum chinense DC.

BMC plant biology·2026

Same author

A mechanistic exploration of IL-1β / p38 MAPK / MMP-2 pathway in inguinal hernia: evidence from clinical specimens and in vitro model.

Surgical endoscopy·2026

Same author

Heterosis unveiled in root-related traits and saikosaponins content between triploid F<sub>1</sub> hybrids and parental <i>Bupleurum chinense</i> DC.

Frontiers in plant science·2026

Same author

Pharmacokinetic Studies of a Novel c-Met Targeting PROTAC Drug Candidate Using UPLC-MS/MS Quantification Methods.

Biopharmaceutics & drug disposition·2026

Same author

Adaptive Transfer Crystal Graph Convolutional Networks for Accelerated Discovery of Organic Perovskite Photovoltaic Materials.

The journal of physical chemistry. A·2026

Same author

Analysis of the potential regulatory mechanism of PIEZO1 in Alzheimer's disease based on RNA sequencing.

IBRO neuroscience reports·2025

Search research articles

Related Experiment Video

Updated: Jul 6, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Transformer guided self-adaptive network for multi-scale skin lesion image segmentation.

Chao Xin¹, Zhifang Liu¹, Yizhao Ma¹

¹The First Affiliated Hospital of Ningbo University, Ningbo, 315211, China.

Computers in Biology and Medicine

|January 7, 2024

Summary

A new self-adaptive position-aware model, SapFormer, enhances skin lesion segmentation by capturing global context and spatial relationships, improving diagnostic accuracy for better patient care.

Keywords:

Segmentation Self-adaptive feature extraction Skin lesion Vision transformer

More Related Videos

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging

Published on: May 5, 2011

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: Jul 6, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging

Quantitative Visualization and Detection of Skin Cancer Using Dynamic Thermal Imaging

Published on: May 5, 2011

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

Area of Science:

Dermatology
Medical Image Analysis
Artificial Intelligence

Background:

Skin lesion diagnosis relies heavily on accurate segmentation.
Traditional CNNs struggle with distant connections and intricate features in skin lesion images.
Existing methods often overlook regional category distribution and pixel spatial relationships.

Purpose of the Study:

To propose SapFormer, a self-adaptive position-aware model for improved skin lesion segmentation.
To enhance capture of global context, fine-grained details, and spatial relationships.
To adapt to diverse positional characteristics and reduce missegmentation of non-lesion areas.

Main Methods:

Developed SapFormer, a multi-scale dynamic position-aware transformer network.
Employed hybrid transformers for multi-scale feature encoding and positional sensing.
Utilized a transformer decoder with a self-adaptive framework and cross-attention for regional feature optimization.

Main Results:

SapFormer achieved high accuracy (97.9%, 94.3%, 95.7%) on ISIC datasets.
Demonstrated superior IOU (93.2%, 86.4%, 89.4%) and DSC (96.4%, 92.6%, 94.3%) metrics compared to SOTA models.
Showcased excellent noise resistance and fine-grained feature extraction in lesion segmentation.

Conclusions:

Transformer-guided position-aware networks significantly boost semantic skin lesion segmentation.
SapFormer's ability to capture spatial relationships and details improves diagnostic efficiency.
The model aids dermatologists in accurate diagnosis, enhancing patient care and clinical decision-making.