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: Apr 15, 2026

Automated Joint Space Detection Improves Bone Segmentation Accuracy
06:45

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

294

An Interpretable Hybrid SFNet Deep Learning Framework for Multi-Site Bone Fracture Detection in Medical Imaging.

Wijdan S Aljebreen1, Da'ad Albahdal1, Shuaa S Alharbi1

  • 1Department of Information Technology, College of Computer, Qassim University, Buraydah 52571, Saudi Arabia.

Diagnostics (Basel, Switzerland)
|April 14, 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

Explainable Lightweight AI for the Identification of Right-Sided Cardiac Dysfunction in a Saudi Arabian Diabetic Cohort.

Journal of clinical medicine·2026
Same author

Advanced Deep Learning Models for Classifying Dental Diseases from Panoramic Radiographs.

Diagnostics (Basel, Switzerland)·2026
Same author

Advances in Image-Based Diagnosis of Diabetic Foot Ulcers Using Deep Learning and Machine Learning: A Systematic Review.

Biomedicines·2025
Same author

Multimodal image fusion for enhanced vehicle identification in intelligent transport.

PeerJ. Computer science·2025
Same author

Neuroimaging and Machine Learning in OCD: Advances in Diagnostic and Therapeutic Insights.

Brain sciences·2025
Same author

Multimodal scene recognition using semantic segmentation and deep learning integration.

PeerJ. Computer science·2025

A new Hybrid Selective Feature Network (Hybrid SFNet) improves multi-site bone fracture detection accuracy and boundary localization. This interpretable AI tool enhances orthopedic diagnosis by reducing errors and aiding clinical decisions.

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Orthopedics

Background:

  • Accurate bone fracture detection is crucial for orthopedic diagnosis and trauma care.
  • Manual interpretation of medical images is time-consuming and prone to inter-observer variability.
  • Subtle or multi-site fractures pose significant diagnostic challenges.

Purpose of the Study:

  • To develop an interpretable Hybrid Selective Feature Network (Hybrid SFNet) for enhanced multi-site bone fracture detection.
  • To improve fracture boundary localization and overall diagnostic performance.
  • To provide a tool that supports clinical decision-making in orthopedic imaging.

Main Methods:

  • The Hybrid SFNet integrates multi-scale convolutional features and a semantic flow mechanism.
Keywords:
artificial intelligencebone fracture detectiondeep learning modelsdiagnosisgrad-CAMhybrid SFNetimage segmentationmachine learningmedical imaging

More Related Videos

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

7.7K

Related Experiment Videos

Last Updated: Apr 15, 2026

Automated Joint Space Detection Improves Bone Segmentation Accuracy
06:45

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

294
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

7.7K
  • Preprocessing included Canny edge detection, normalization, and data augmentation.
  • Gradient-weighted Class Activation Mapping (Grad-CAM) was used for model interpretability.

    • Main Results:

    • The model achieved 90% accuracy, 94% precision, 77% recall, and 85% F1-score for fractured cases.
    • Class-weighted loss improved recall to 85%, reducing false negatives by approximately 35%.
    • Cohen's Kappa reached 0.79 and MCC reached 0.76 with weighted configuration.

    Conclusions:

    • The Hybrid SFNet offers an interpretable and effective solution for multi-site bone fracture detection.
    • Multi-scale features and semantic flow enhance detection and boundary delineation.
    • The model shows potential for clinical decision support in orthopedic imaging.