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: Aug 16, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

603

Human Fall Detection Using 3D Multi-Stream Convolutional Neural Networks with Fusion.

Thamer Alanazi1, Ghulam Muhammad1

  • 1Department of Computer Engineering, College of Computer and Information Sciences, King Saud University, Riyadh 11543, Saudi Arabia.

Diagnostics (Basel, Switzerland)
|December 23, 2022
PubMed
Summary

This study introduces an advanced automatic human fall detection system using a novel multi-stream convolutional neural network (CNN). The system accurately identifies falls in elderly individuals, enabling faster medical response and improving healthcare outcomes.

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

A Systematic Review of Diffusion Models for Medical Image-Based Diagnosis: Methods, Taxonomies, Clinical Integration, Explainability, and Future Directions.

Diagnostics (Basel, Switzerland)·2026
Same author

Internet of medical things and blockchain-enabled patient-centric agent through SDN for remote patient monitoring in 5G network.

Scientific reports·2024
Same author

Potent Bile Acid Microbial Metabolites Modulate <i>Clostridium perfringens</i> Virulence.

Pathogens (Basel, Switzerland)·2023
Same author

A Novel Deep Learning-Based Classification Framework for COVID-19 Assisted with Weighted Average Ensemble Modeling.

Diagnostics (Basel, Switzerland)·2023
Same author

Metal organic framework derived NiO <sub></sub> nanoparticles for application as a hole transport layer in perovskite solar cells.

RSC advances·2023
Same author

Characteristics, Management, and Outcomes of Community-Acquired Pneumonia due to Respiratory Syncytial Virus: A Retrospective Study.

Pulmonary medicine·2023

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Biomedical Engineering

Background:

  • Human falls, particularly among the elderly, result in severe injuries and disability, posing a significant public health challenge.
  • Accurate and timely fall detection is crucial for immediate medical intervention and improved patient outcomes.
  • Vision-based technologies, especially deep learning, show promise for reliable fall action recognition.

Purpose of the Study:

  • To develop an automatic human fall detection system utilizing multi-stream convolutional neural networks (CNNs) with image fusion.
  • To enhance the accuracy and efficiency of fall detection for elderly individuals.
  • To provide a robust solution for real-time fall incident detection and reporting.

Main Methods:

  • A multi-level image-fusion approach was employed, processing 16 frames of video to highlight movement differences.
Keywords:
3D-CNN and 2D-CNNconvolution neural networksdeep learningfusion networkshuman fall detection

More Related Videos

Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

10.8K
Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction
06:19

Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction

Published on: August 16, 2024

481

Related Experiment Videos

Last Updated: Aug 16, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

603
Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

10.8K
Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction
06:19

Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction

Published on: August 16, 2024

481
  • A novel, lightweight multi-stream CNN model (4S-3DCNN) with a four-branch architecture was designed for fall classification.
  • The system was evaluated on the Le2i fall detection dataset, comprising over 6392 video sequences.

    • Main Results:

    • The proposed 4S-3DCNN model achieved high performance metrics: 99.03% accuracy, 99.00% sensitivity, 99.68% specificity, and 99.00% precision.
    • Three-fold cross-validation was used to ensure model generalization and prevent overfitting.
    • The system demonstrated superior performance compared to state-of-the-art models like GoogleNet, SqueezeNet, ResNet18, and DarkNet19.

    Conclusions:

    • The developed multi-stream CNN fall detection system is highly accurate and effective.
    • This technology has the potential to significantly improve emergency response for fall incidents, especially in the elderly population.
    • The proposed model represents a significant advancement in vision-based fall detection systems.