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 Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Development and validation of a regression equation for VO2 peak prediction in patients with heart and neurologic diseases.

Medicine·2026
Same author

Upregulated Expression of TRPV1 and TRPV4 in the Urethra Following Cyclophosphamide-Induced Cystitis in Rats: A Potential Mechanism of Bladder-Urethral Dysfunction.

International neurourology journal·2026
Same author

On-site detection of airborne foodborne pathogens using a field-deployable recombinase polymerase amplification and CRISPR/Cas12a cleavage activity assay.

Biosensors & bioelectronics·2026
Same author

A Modular Vaccine Platform Against SARS-CoV-2 Based on Self-Assembled Protein Nanoparticles.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
Same author

Feasibility and Preliminary Effectiveness of a Mobile App-Based Personalized Exercise Program in Older Patients With Chronic Knee Osteoarthritis: Pilot Randomized Controlled Trial.

JMIR mHealth and uHealth·2025
Same author

Cohort study of older adults receiving home-based primary care in South Korea: cohort profile.

BMJ open·2025

Related Experiment Video

Updated: Jul 5, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

3.8K

Investigating Activity Recognition for Hemiparetic Stroke Patients Using Wearable Sensors: A Deep Learning Approach

Youngmin Oh1, Sol-A Choi2, Yumi Shin2

  • 1School of Computing, Gachon University, Seongnam 13120, Republic of Korea.

Sensors (Basel, Switzerland)
|January 11, 2024
PubMed
Summary

Classifying arm movements in stroke survivors using wearable sensors is key for home rehabilitation. Jointly training models with data from stroke patients and non-disabled individuals significantly improves accuracy.

Keywords:
activities of daily livingclassificationdeep learninghemiparesishuman action recognitionrange of motionstroke rehabilitationupper extremity

More Related Videos

A Human-machine-interface Integrating Low-cost Sensors with a Neuromuscular Electrical Stimulation System for Post-stroke Balance Rehabilitation
11:06

A Human-machine-interface Integrating Low-cost Sensors with a Neuromuscular Electrical Stimulation System for Post-stroke Balance Rehabilitation

Published on: April 12, 2016

10.5K
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
08:15

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

494

Related Experiment Videos

Last Updated: Jul 5, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

3.8K
A Human-machine-interface Integrating Low-cost Sensors with a Neuromuscular Electrical Stimulation System for Post-stroke Balance Rehabilitation
11:06

A Human-machine-interface Integrating Low-cost Sensors with a Neuromuscular Electrical Stimulation System for Post-stroke Balance Rehabilitation

Published on: April 12, 2016

10.5K
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
08:15

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

494

Area of Science:

  • Biomedical Engineering
  • Rehabilitation Technology
  • Machine Learning in Healthcare

Background:

  • Assessing daily limb use post-stroke is vital for effective rehabilitation.
  • Wearable sensors offer non-intrusive methods for movement classification in home-based stroke recovery systems.
  • Stroke patient movement data presents challenges like variability and sparsity for classification models.

Purpose of the Study:

  • To develop and evaluate a movement classification system for hemiparetic stroke patients using wearable sensors.
  • To investigate the impact of training data composition (stroke patients vs. non-disabled individuals) on model performance.
  • To assess the effectiveness of data augmentation and the influence of movement asymmetry on classification accuracy.

Main Methods:

  • Collected movement data from 15 hemiparetic stroke patients and 29 non-disabled individuals performing range of motion and activities of daily living tasks.
  • Utilized five inertial measurement units worn by participants in a home environment.
  • Trained and evaluated a 1D convolutional neural network using ND-only, Stroke-only, and combined ND + Stroke training datasets.
  • Applied data augmentation techniques (axis rotation) and analyzed performance based on movement symmetry.

Main Results:

  • Joint training with both non-disabled and stroke data significantly improved the F1-score by 31.6% (vs. ND-only) and 10.6% (vs. Stroke-only).
  • Data augmentation enhanced F1-scores by an average of 11.3% across all training conditions.
  • Movement asymmetry in the Stroke group reduced the F1-score by 25.9% compared to symmetric movements.

Conclusions:

  • Combining data from non-disabled individuals and stroke patients enhances movement classification accuracy for hemiparetic stroke rehabilitation.
  • Data augmentation and accounting for movement asymmetry are crucial for optimizing wearable sensor-based rehabilitation systems.
  • The findings support the development of intelligent home-based rehabilitation systems for stroke recovery.