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: Dec 8, 2025

Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

11.0K

Wearables and Deep Learning Classify Fall Risk From Gait in Multiple Sclerosis.

Brett M Meyer, Lindsey J Tulipani, Reed D Gurchiek

    IEEE Journal of Biomedical and Health Informatics
    |September 18, 2020
    PubMed
    Summary

    Wearable sensors and deep learning can identify persons with multiple sclerosis (PwMS) who have fallen. This technology offers a simple, inexpensive method for fall risk assessment in PwMS.

    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

    Interleukin 6 Receptor Blockade for Relapse Prevention in Myelin Oligodendrocyte Glycoprotein Antibody-Associated Disease.

    JAMA neurology·2026
    Same author

    Challenges and future directions for multiple sclerosis after the 2024 McDonald diagnostic criteria.

    Nature medicine·2026
    Same author

    Diagnostic delay in patients with myelin oligodendrocyte antibody-associated disease presenting with elevated intracranial pressure.

    Multiple sclerosis and related disorders·2026
    Same author

    Global Access to Diagnostic Paraclinical Testing Incorporated in the 2024 Revised McDonald Criteria: Disparities and Opportunity.

    Neurology. Clinical practice·2026
    Same author

    Application of the 2024 McDonald Criteria in Individuals With Nonspecific Symptoms or Incidental Imaging Findings in a Multicenter Study.

    Neurology·2026
    Same author

    Reducing Video Verification Burden: Machine Learning Classification of Head Acceleration Events in Youth Football.

    Research square·2026

    Area of Science:

    • Neurology
    • Biomedical Engineering
    • Data Science

    Background:

    • Falls are a major concern for persons with multiple sclerosis (PwMS), often leading to interventions only after a fall occurs.
    • Objective fall risk assessments are needed for proactive intervention, but current methods are often laboratory-based and not clinically deployable.
    • Previous research identified gait biomechanics differences in PwMS who have fallen, but these haven't been used for detection with wearable technology.

    Purpose of the Study:

    • To develop and validate a fall status classification method for PwMS using wearable sensor data.
    • To leverage deep learning, specifically a bidirectional long short-term memory (BiLSTM) network, for fall detection.
    • To compare the performance of the BiLSTM model against traditional machine learning and clinical measures.

    More Related Videos

    Author Spotlight: Innovations in iTUG Test for Enhanced Risk Assessment and Cognitive Insights
    05:26

    Author Spotlight: Innovations in iTUG Test for Enhanced Risk Assessment and Cognitive Insights

    Published on: October 25, 2024

    1.5K
    Home-Based Monitor for Gait and Activity Analysis
    07:24

    Home-Based Monitor for Gait and Activity Analysis

    Published on: August 8, 2019

    7.1K

    Related Experiment Videos

    Last Updated: Dec 8, 2025

    Design and Analysis for Fall Detection System Simplification
    08:05

    Design and Analysis for Fall Detection System Simplification

    Published on: April 6, 2020

    11.0K
    Author Spotlight: Innovations in iTUG Test for Enhanced Risk Assessment and Cognitive Insights
    05:26

    Author Spotlight: Innovations in iTUG Test for Enhanced Risk Assessment and Cognitive Insights

    Published on: October 25, 2024

    1.5K
    Home-Based Monitor for Gait and Activity Analysis
    07:24

    Home-Based Monitor for Gait and Activity Analysis

    Published on: August 8, 2019

    7.1K

    Main Methods:

    • Utilized accelerometer data from two wearable sensors during a one-minute walking task.
    • Developed a bidirectional long short-term memory (BiLSTM) deep neural network for retrospective fall status classification.
    • Compared the BiLSTM model's performance (AUC) against models trained on spatiotemporal gait parameters, statistical features from sensor data, and patient/neurologist-reported measures.

    Main Results:

    • The BiLSTM model achieved an Area Under the Curve (AUC) of 0.88 in identifying PwMS who have recently fallen.
    • This deep learning approach significantly outperformed other methods, showing AUC improvements of 21% over spatiotemporal gait parameters, 16% over statistical sensor features, 19% over patient-reported measures, and 24% over neurologist-administered measures.
    • The method demonstrated good performance with minimal data requirements (one minute of walking from two sensors).

    Conclusions:

    • A BiLSTM deep neural network utilizing accelerometer data from wearable sensors can effectively identify PwMS who have fallen.
    • This approach offers a promising, simple, and inexpensive method for fall risk assessment in clinical settings.
    • The findings support the use of wearable sensors for objective and accessible fall risk monitoring in PwMS.