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: Sep 20, 2025

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

A pre-impact fall detection data segmentation method based on multi-channel convolutional neural network and class

Mingxu Feng1,2, Jizhong Liu1

  • 1The Nanchang Key Laboratory of Medical and Technology Research, Nanchang University, Nanchang 330031, People's Republic of China.

Physiological Measurement
|June 10, 2022
PubMed
Summary
This summary is machine-generated.

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

Decoding multi-class motor attempt from the affected unilateral limbs in chronic stroke patients.

Journal of neuroengineering and rehabilitation·2026
Same author

Cu-MOF derived porous carbon and nitrogen-doped longan shell carbon modified electrochemical sensor for rapid detection of enrofloxacin.

Mikrochimica acta·2026
Same author

Cortical Response and Functional Connectivity in Patients With Subacute Stroke During Dual Task: An fNIRS Study.

Archives of physical medicine and rehabilitation·2026
Same author

In Situ Grown Cs<sub>2</sub>Ag<sub>0.3</sub>Na<sub>0.7</sub>InCl<sub>6</sub> Double Perovskites within a PVDF Matrix for High-Performance Flexible Piezoelectric Nanogenerators.

ACS applied materials & interfaces·2025
Same author

The search for life signatures on Mars by the Tianwen-3 Mars sample return mission.

National science review·2024
Same author

A lightweight deep-learning model for parasite egg detection in microscopy images.

Parasites & vectors·2024
Same journal

Continuous tracking of aortic aneurysm diameter with peripheral pulse waves: a computational framework combining sequential Markov chain Monte Carlo with Kalman filtering.

Physiological measurement·2026
Same journal

The 2026 global roadmap for textile-integrated wearable technologies in health.

Physiological measurement·2026
Same journal

Augmenting single-lead ECG interpretation through QRS waveform decomposition and rotation.

Physiological measurement·2026
Same journal

Dynamic Beat-to-Beat Blood Pressure Estimation using a Multi-modal Wearable Deep Learning Approach.

Physiological measurement·2026
Same journal

Dual warm-start fusion versus attention-based fusion in low-label ECG-PCG classification: a controlled ablation study.

Physiological measurement·2026
Same journal

Inter-patient multi-label ECG classification via low-rank adaptation fine-tuned large language models with dynamic graph convolutional network.

Physiological measurement·2026
See all related articles

A novel segmentation method using a convolutional neural network (CNN) and class activation mapping accurately segments pre-impact fall detection data. This approach ensures real-time performance and enhances machine learning accuracy for wearable devices.

Area of Science:

  • Biomedical Engineering
  • Signal Processing
  • Machine Learning

Background:

  • Accurate pre-impact fall detection is crucial for timely intervention.
  • Continuous inertial sensor data presents challenges for efficient fall detection segmentation.
  • Existing methods may lack the precision and real-time capabilities required for wearable systems.

Purpose of the Study:

  • To investigate a data segmentation method for pre-impact fall detection.
  • To partition critical data segments from continuous inertial sensor data for improved classification.
  • To develop a trigger-based algorithm for efficient and accurate data segmentation.

Main Methods:

  • A multi-channel convolutional neural network (CNN) combined with class activation mapping was employed.
Keywords:
class activity mappingconvolutional neural networksfall detectionmulti-channelpre-impact

More Related Videos

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

652

Related Experiment Videos

Last Updated: Sep 20, 2025

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
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

652
  • A pre-impact fall detection dataset was created, segmenting fall data from peak acceleration and daily activities cyclically.
  • A data segmentation strategy was derived from heat maps, fall characteristics, and real-time constraints.

    • Main Results:

    • The proposed method achieved real-time performance for pre-impact fall detection.
    • The segmentation strategy, initiated at acceleration magnitude < 9.217 m/s², uses a 325 ms window.
    • Machine learning algorithms validated the strategy, exceeding 94.8% accuracy.

    Conclusions:

    • The method automatically identifies relevant data segments, optimizing computational resources on wearable devices.
    • Reduced segmentation window duration enhances the real-time performance of pre-impact fall detection.
    • This approach is adaptable to various machine learning algorithms, improving overall fall detection systems.