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: May 14, 2026

Automatic Surgery in Transcatheter Aortic Valve Replacement Using Augmented Reality
07:46

Automatic Surgery in Transcatheter Aortic Valve Replacement Using Augmented Reality

Published on: August 9, 2024

Monitoring Adherence to Post-Sternotomy Movement Precautions: A Computer Vision and Generative AI Approach.

Davide Ferrari1, Giulia Besana1, Anna Dotti1

  • 1Department of Management, Information and Production Engineering, University of Bergamo, Dalmine (Bergamo), Italy.

Studies in Health Technology and Informatics
|May 12, 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

Data processing pipeline for managing enteric methane emissions from sniffer: Application on dairy farming systems.

JDS communications·2026
Same author

Comparison of Proteomic Analysis of Cerebrospinal Fluid From Neurological Patients With and Without Amyotrophic Lateral Sclerosis.

Journal of neurochemistry·2026
Same author

Rethinking hypercortisolism screening in obesity and diabetes: A systematic review and meta-analysis.

Metabolism: clinical and experimental·2026
Same author

Skin Application of Inflammasome Inhibitor MCC950 Prevents Psoriasis-like Cutaneous Manifestations and Associated Systemic Inflammation.

Inflammation·2026
Same author

Dyslipidemias associated with endocrine disorders: a position statement of the working group of the nutrition hormones and metabolism club of the italian society of endocrinology (SIE).

Journal of endocrinological investigation·2026
Same author

Sex differences in classic congenital adrenal hyperplasia: a multicenter, real-world analysis.

Frontiers in endocrinology·2026

This study developed a computer vision system to monitor post-sternotomy movement compliance. Generative AI improved the system, enabling accurate, real-time patient monitoring for better sternal healing.

Area of Science:

  • Biomedical Engineering
  • Computer Vision
  • Artificial Intelligence

Background:

  • Adherence to post-sternotomy movement precautions is crucial for sternal healing.
  • Patients frequently struggle to consistently follow these essential movement restrictions.
  • Ensuring compliance is vital for optimal recovery after sternotomy.

Purpose of the Study:

  • To develop a real-time computer vision system for monitoring adherence to post-sternotomy precautions.
  • To evaluate the effectiveness of AI-generated data in training object detection models for this application.
  • To create a non-invasive and low-cost solution for patient monitoring.

Main Methods:

  • A real-time computer vision system was designed using the YOLOv11 object detection model.
Keywords:
Cardiac RehabilitationGenerative Artificial IntelligenceSternotomy

More Related Videos

Web-based Clinician Guide to Record Compatible Video of Standardized Drinking Task Kinematics for Computer Vision Analysis
07:28

Web-based Clinician Guide to Record Compatible Video of Standardized Drinking Task Kinematics for Computer Vision Analysis

Published on: November 28, 2025

Related Experiment Videos

Last Updated: May 14, 2026

Automatic Surgery in Transcatheter Aortic Valve Replacement Using Augmented Reality
07:46

Automatic Surgery in Transcatheter Aortic Valve Replacement Using Augmented Reality

Published on: August 9, 2024

Web-based Clinician Guide to Record Compatible Video of Standardized Drinking Task Kinematics for Computer Vision Analysis
07:28

Web-based Clinician Guide to Record Compatible Video of Standardized Drinking Task Kinematics for Computer Vision Analysis

Published on: November 28, 2025

  • The YOLOv11 model was trained on a dataset combining real and AI-generated images.
  • A secondary monitoring algorithm processed YOLO inference results to classify actions, not just single frames.

    • Main Results:

    • The integration of synthetic data boosted YOLO model performance, achieving a mAP50 of 80.3%.
    • The monitoring algorithm achieved 85% accuracy in classifying non-compliant actions during real-time validation.
    • No compliant actions were misclassified as non-compliant, indicating high specificity.

    Conclusions:

    • The study demonstrates a feasible, low-cost, non-invasive method for monitoring post-sternotomy precautions.
    • Generative AI effectively addressed data scarcity challenges in training the computer vision model.
    • This technology holds potential for improving patient adherence and surgical recovery outcomes.