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

Wearable Camera-Based Dietary Assessment of Mother-Father Dyads in Urban and Rural Households in Ghana.

Current developments in nutrition·2026
Same author

Eating architecture components and their associations with BMI in urban and rural Ghanaian mothers, fathers, children, and adolescents, assessed using a wearable camera: A cross-sectional study.

Chronobiology international·2026
Same author

Plasma inflammatory biomarker profiles across the Alzheimer's disease spectrum in the Bio-Hermes cohort.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2026
Same author

Antipsychotic-induced weight gain in psychosis: causal mediation analysis and feasibility study of causal actionable prediction model development using counterfactuals to target obesity.

The British journal of psychiatry : the journal of mental science·2026
Same author

Predicting cardiopulmonary exercise testing outcomes in congenital heart disease through multimodal data integration and geometric learning.

Scientific reports·2026
Same author

Digitization and Linkage of PDF Formatted 12-Lead Electrocardiograms in Adult Congenital Heart Disease.

CJC pediatric and congenital heart disease·2026

Related Experiment Video

Updated: Dec 13, 2025

Asymmetric Walkway: A Novel Behavioral Assay for Studying Asymmetric Locomotion
08:19

Asymmetric Walkway: A Novel Behavioral Assay for Studying Asymmetric Locomotion

Published on: January 15, 2016

9.2K

Cross-Subject and Cross-Modal Transfer for Generalized Abnormal Gait Pattern Recognition.

Xiao Gu, Yao Guo, Fani Deligianni

    IEEE Transactions on Neural Networks and Learning Systems
    |July 30, 2020
    PubMed
    Summary

    This study introduces a novel deep learning approach for recognizing abnormal gaits by effectively transferring knowledge across different data types and subjects. The method enhances recognition accuracy, even with noisy sensor data, improving generalization for gait analysis.

    More Related Videos

    Clinical-oriented Three-dimensional Gait Analysis Method for Evaluating Gait Disorder
    06:54

    Clinical-oriented Three-dimensional Gait Analysis Method for Evaluating Gait Disorder

    Published on: March 4, 2018

    14.5K
    Cross-Modal Multivariate Pattern Analysis
    13:51

    Cross-Modal Multivariate Pattern Analysis

    Published on: November 9, 2011

    20.3K

    Related Experiment Videos

    Last Updated: Dec 13, 2025

    Asymmetric Walkway: A Novel Behavioral Assay for Studying Asymmetric Locomotion
    08:19

    Asymmetric Walkway: A Novel Behavioral Assay for Studying Asymmetric Locomotion

    Published on: January 15, 2016

    9.2K
    Clinical-oriented Three-dimensional Gait Analysis Method for Evaluating Gait Disorder
    06:54

    Clinical-oriented Three-dimensional Gait Analysis Method for Evaluating Gait Disorder

    Published on: March 4, 2018

    14.5K
    Cross-Modal Multivariate Pattern Analysis
    13:51

    Cross-Modal Multivariate Pattern Analysis

    Published on: November 9, 2011

    20.3K

    Area of Science:

    • Biomedical Engineering
    • Computer Science
    • Machine Learning

    Background:

    • Abnormal gait recognition is challenging due to interleaved pattern-specific and subject-specific features, leading to model overfitting and poor generalization.
    • Limited availability of precise Motion Capture (Mocap) data and noise in data from wearable or vision sensors hinder effective abnormal gait analysis.
    • Existing methods struggle to generalize to new subjects and effectively utilize data from diverse sources.

    Purpose of the Study:

    • To develop a robust deep learning framework for abnormal gait recognition that overcomes data limitations and sensor noise.
    • To enable effective cross-modal and cross-subject knowledge transfer for improved gait analysis.
    • To enhance the generalization capability of abnormal gait recognition models.

    Main Methods:

    • A cascade of deep architectures employing cross-modal transfer to map noisy sensor data (RGBD, wearable) to accurate 4-D Mocap representations.
    • Cross-subject transfer using a multi-encoder autoencoder architecture to disentangle subject-specific and abnormal pattern-specific gait features.
    • Validation using multimodal gait data from a multicamera Mocap system, synchronized electromyography (EMG), and RGBD camera 4-D skeleton data.

    Main Results:

    • Significant improvements in classification accuracy were achieved for abnormal gait recognition across both Mocap and noisy modalities.
    • The proposed cross-modal and cross-subject transfer learning approach effectively addressed the challenges of limited precise data and sensor noise.
    • The disentanglement of gait features led to more generalizable and accurate abnormal gait recognition models.

    Conclusions:

    • The developed cascade of deep architectures provides a powerful solution for abnormal gait recognition, particularly in scenarios with limited high-quality data.
    • Cross-modal and cross-subject transfer learning are effective strategies for enhancing the performance and generalizability of gait analysis systems.
    • This methodology holds promise for improving the diagnosis and monitoring of gait abnormalities using diverse and potentially noisy sensor data.