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: Jan 11, 2026

Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

11.1K

Estimation of Fugl-Meyer Assessment Upper-Extremity Sub-Scores Using a Mixup-Augmented LSTM Autoencoder and Wearable

Minghao Liu1, Hsuan-Yu Lu2, Shuk-Fan Tong1

  • 1Department of Biomedical Engineering, The Chinese University of Hong Kong, Hong Kong SAR, China.

Sensors (Basel, Switzerland)
|November 13, 2025
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

A Portable Soft Robotic Glove with Fully Functional Thumb Assistance for Complex Dexterous Fine Motor Skills.

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

CIITA/PRMT5 promote CD4<sup>+</sup>Gzma<sup>+</sup> T cell activation via H3R2me2-mediated endothelial expression of MHC class II in smoking-induced atherosclerosis.

BMC medicine·2026
Same author

Effect of Calcareous Material Particle Size via Separate Grinding on the Burnability and Microstructure Development of Portland Cement Clinker.

Materials (Basel, Switzerland)·2026
Same author

Prompting Rain Off: Evolving Compact Dual Prompts for Continual De-Raining.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same author

Construction and Application of a Tactile Somatosensory Comfort Model for Scrubbing Tasks.

Biomimetics (Basel, Switzerland)·2026
Same author

Identification of herbal formula Huatanqushihuoxue formula as a potential therapeutic agent for metabolism-related fatty liver disease: a multi-omics and network pharmacology approach.

Frontiers in immunology·2026

This study developed a deep learning model using wearable sensors to automatically estimate Fugl-Meyer Assessment for the Upper Extremity (FMA-UE) scores. This innovation offers a faster, more accessible method for assessing motor function after stroke.

Area of Science:

  • Biomedical Engineering
  • Rehabilitation Science
  • Artificial Intelligence in Healthcare

Background:

  • Stroke significantly impacts motor function, necessitating accurate assessment for rehabilitation.
  • The Fugl-Meyer Assessment for the Upper Extremity (FMA-UE) is a standard but time-consuming clinical tool.
  • Current remote assessment methods using sensors and AI face challenges in detailed motor function scoring.

Purpose of the Study:

  • To introduce a deep learning framework for automated estimation of FMA-UE total and subdivision scores.
  • To enable more frequent and accessible motor function assessments for stroke survivors.
  • To overcome limitations of traditional FMA-UE assessments in terms of time and expertise.

Main Methods:

  • Collected data from 15 participants using four inertial measurement units (IMUs) on the arm and trunk.
Keywords:
Fugl–Meyer assessmentIMUdeep learningremote assessmentstroke

More Related Videos

Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable
09:24

Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable

Published on: May 17, 2024

2.1K
A Method for Quantifying Upper Limb Performance in Daily Life Using Accelerometers
07:24

A Method for Quantifying Upper Limb Performance in Daily Life Using Accelerometers

Published on: April 21, 2017

13.0K

Related Experiment Videos

Last Updated: Jan 11, 2026

Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

11.1K
Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable
09:24

Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable

Published on: May 17, 2024

2.1K
A Method for Quantifying Upper Limb Performance in Daily Life Using Accelerometers
07:24

A Method for Quantifying Upper Limb Performance in Daily Life Using Accelerometers

Published on: April 21, 2017

13.0K
  • Developed a deep learning model integrating an LSTM-based autoencoder and mixup augmentation.
  • Utilized a leave-one-subject-out cross-validation (LOSOCV) for robust performance evaluation.

    • Main Results:

    • The model achieved R2 values > 0.82 and Pearson's r > 0.90 for all FMA-UE subparts (A-D).
    • Normalized root-mean-square errors (NRMSE) were < 0.14 for subparts and 0.0678 for the total FMA-UE score.
    • Demonstrated strong generalization and robustness in estimating detailed motor function scores.

    Conclusions:

    • A concise, sensor-based assessment framework can reliably predict detailed FMA-UE motor function scores.
    • This approach enhances the accessibility and efficiency of stroke motor function assessment.
    • The developed deep learning model shows significant potential for remote and automated rehabilitation monitoring.