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: Jul 19, 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

ERYXSeg: a hybrid CNN architecture for robust and resource-aware wound segmentation.

Trishaani Acharjee1, Rajdeep Chatterjee1, Mahendra Kumar Gourisaria1

  • 1School of Computer Engineering, KIIT Deemed to be University, Bhubaneswar, India.

Frontiers in Surgery
|July 6, 2026
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

Speech impairment detection in children using time frequency features of speech and deep learning techniques.

Frontiers in human neuroscience·2026
Same author

A real-time industrial safety automation using YOLO architectures leveraging diverse chromatic domains.

Scientific reports·2026
Same author

Cervical cancer classification using a novel hybrid approach.

Frontiers in oncology·2025
Same author

Performance optimization of ethanol blends in diesel model using Taguchi and grey relational approach.

Scientific reports·2025
Same author

A scalable framework for evaluating multiple language models through cross-domain generation and hallucination detection.

Scientific reports·2025
Same author

A systematic literature review on the use of multicriteria decision making methods for small and medium-sized enterprises innovation assessment.

Frontiers in artificial intelligence·2025

ERYXSeg, a new deep learning model, accurately segments various wounds for better treatment planning. This efficient architecture achieves state-of-the-art results, enabling real-time clinical use.

Area of Science:

  • Medical image analysis
  • Artificial intelligence in healthcare
  • Computational pathology

Background:

  • Accurate wound segmentation is crucial for data-driven treatment planning and healing assessment.
  • Existing methods may lack robustness across diverse wound types.

Purpose of the Study:

  • To introduce ERYXSeg, an advanced deep learning architecture for precise wound segmentation.
  • To evaluate ERYXSeg's performance against state-of-the-art models on multiple wound datasets.

Main Methods:

  • Developed ERYXSeg, integrating residual skip connections, a parameter-efficient core, and a boundary-aware segmentation head.
  • Trained ERYXSeg from scratch on mixed wound datasets, including foot ulcers.
  • Conducted systematic ablation studies to validate architectural components like attention-gated skip connections and MBConv blocks.
Keywords:
deep learningfoot ulcerimage segmentationmedical imagingwounds

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

Related Experiment Videos

Last Updated: Jul 19, 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

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Main Results:

  • ERYXSeg achieved state-of-the-art performance, outperforming existing models on mixed wound datasets.
  • Highest reported IoU (0.7633) and Dice (0.8658) scores on the foot ulcer dataset.
  • Achieved 0.6910 IoU and 0.8173 Dice scores on the curated mixed wound dataset, demonstrating generalization.

Conclusions:

  • ERYXSeg demonstrates high accuracy and computational efficiency for wound segmentation.
  • The model's robust performance and generalization capabilities make it suitable for real-time clinical deployment.
  • Attention-gated skip connections and MBConv blocks are vital for ERYXSeg's effectiveness.