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 Videos

Identification of Peristomal Skin Alterations Using Convolutional Artificial Neural Networks.

Isabel María López-Medina1, César Hueso-Montoro1, Francisco Charte-Ojeda2

  • 1Department of Nursing, University of Jaén, Jaén, Spain, ujaen.es.

Nursing Research and Practice
|June 17, 2026
PubMed
Summary

Related Concept Videos

Ostomy Care01:24

Ostomy Care

Introduction
An ostomy is a surgical procedure that creates an artificial opening from the intestines to the outside of the body, allowing for the rerouting of effluent. This opening is known as a stoma. A stoma usually protrudes above the skin surface, appearing pink or red, moist, and round, and it lacks nerve sensations.
There are different types of ostomies, including colostomies, ileostomies, and urostomies:

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Factors Associated With the Resolution of Telephone Follow-up in Palliative Home Care Patients of a Public Telemedicine Service: A Retrospective Observational Study Using Machine Learning.

SAGE open nursing·2026
Same author

Environmental Competencies in Nurses and Undergraduate Nursing Students Related to the Effects of Climate Change on Older People's Health.

Nursing reports (Pavia, Italy)·2026
Same author

Self-Efficacy in Palliative Care Among Nursing Professionals: A Mixed-Methods Study.

Journal of clinical nursing·2026
Same author

Influence of Sociodemographic Factors and Incontinence Care Practices on the Development of Incontinence-Associated Dermatitis.

Journal of clinical medicine·2026
Same author

Coping with Death Among Nurses in Ecuador: A Mixed-Methods Study.

Healthcare (Basel, Switzerland)·2026
Same author

Effectiveness of Zinc Oxide Ointments Versus Non-Irritating Barrier Films in the Prevention of Incontinence-Associated Dermatitis.

Medical sciences (Basel, Switzerland)·2026
This summary is machine-generated.

Artificial intelligence models using convolutional neural networks (CNNs) can accurately detect peristomal skin complications in ostomy patients. This AI-driven approach aids in early detection and remote care, improving patient outcomes and reducing healthcare costs.

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Dermatology

Background:

  • Peristomal skin complications (PSCs) are prevalent in ostomy patients, impacting quality of life and increasing healthcare expenses.
  • Current diagnostic methods for PSCs are limited, and AI applications for their treatment are underdeveloped.

Purpose of the Study:

  • To develop and validate preliminary convolutional neural network (CNN) models for binary classification of peristomal skin.
  • To distinguish between healthy peristomal tissue and skin lesions, establishing a foundation for automated diagnostic systems.

Main Methods:

  • A prospective study involving 24 stoma nurses from 17 hospitals collected 1165 peristomal skin images.
  • State-of-the-art pretrained CNNs were utilized for image classification.
Keywords:
artificial intelligenceconvolutional neural networksperistomal skinstoma nurse

Related Experiment Videos

  • Models were evaluated using accuracy, F1-score, and area under the ROC curve, with Grad-Cam for explainability.

    • Main Results:

    • In a standard data split, the best model achieved 0.889 accuracy, 0.890 F1-score, and 0.924 ROC AUC.
    • With patient data leakage prevented, the best model yielded 0.778 accuracy, 0.868 F1-score, and 0.653 ROC AUC.
    • Multiple CNN models were tested, demonstrating robust performance in detecting peristomal skin alterations.

    Conclusions:

    • Developed robust and reliable preliminary AI models for detecting peristomal skin alterations from images.
    • These models enable automatic detection of peristomal skin involvement, facilitating remote care and faster treatment initiation.
    • The AI model represents a significant advancement in ostomy care, promoting early detection, complication prevention, and cost savings.