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: Jul 10, 2026

Deep-Learning Based Multi-Joint Synchronous Tracking for Objective Quantification of Hindlimb Locomotor Kinematics in Rats
06:52

Deep-Learning Based Multi-Joint Synchronous Tracking for Objective Quantification of Hindlimb Locomotor Kinematics in Rats

Published on: April 3, 2026

Classify head-turning in walking: Multi-segment IMU sensing with recurrent neural networks.

Sheng-Ming Hsu1, Jing Nong Liang2, Yun-Ju Lee1

  • 1Department of Industrial Engineering and Engineering Management, National Tsing-Hua University, Hsinchu, Taiwan.

Gait & Posture
|July 8, 2026
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

Non-invasive brain stimulation for enhancing dual-task performance in individuals with chronic poststroke hemiparesis: protocol for systematic review and meta-analysis.

BMJ neurology open·2026
Same author

Task-Specific Reliability and Measurement Error of Frontal Plane Kinematics in Individuals with Patellofemoral Pain: A Preliminary Study.

Methods and protocols·2026
Same author

Developmental system drift in dorsoventral patterning is linked to transitions to autonomous development in Annelida.

Nature communications·2026
Same author

Recognizing intentions with body segmental cues of gait cycles before direction change during continuous walking.

Gait & posture·2025
Same author

Transarterial embolization outperforms radiofrequency ablation for thyroid goiters exceeding 100 mL: a study on efficacy and safety.

European radiology·2025
Same author

Comparison of gluteal muscle central activation in individuals with and without patellofemoral pain.

Frontiers in physiology·2025

Wearable sensors accurately detect head turns during walking, enabling predictive gait analysis and adaptive systems. This research enhances interactive technologies by analyzing head movement patterns.

Area of Science:

  • Biomechanics and Human Movement
  • Wearable Technology and Sensor Fusion
  • Machine Learning for Gait Analysis

Background:

  • Head movement is crucial for gait control and orientation during locomotion.
  • Early detection of head turns during walking is vital for advanced interactive systems.
  • Wearable inertial measurement units (IMUs) offer real-time movement inference capabilities.

Purpose of the Study:

  • To assess the reliability of multi-segment IMU data in detecting and classifying head turns and their direction during walking.
  • To hypothesize that neural networks can capture temporal integration patterns of pre-movement and active head turning.
  • To optimize sensor placement and data processing for head turn prediction.

Main Methods:

  • Fifty participants walked while wearing eight IMUs on the head, chest, wrist, lower back, heels, and toes.
Keywords:
Body segment sensingDeep learningDynamic movement intentionHead motion detectionInertial measurement unitStatus recognition

Related Experiment Videos

Last Updated: Jul 10, 2026

Deep-Learning Based Multi-Joint Synchronous Tracking for Objective Quantification of Hindlimb Locomotor Kinematics in Rats
06:52

Deep-Learning Based Multi-Joint Synchronous Tracking for Objective Quantification of Hindlimb Locomotor Kinematics in Rats

Published on: April 3, 2026

  • IMU signals were segmented and processed using a sliding-window approach across pre-turn and during-turn phases.
  • Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks were trained for four-class classification.

    • Main Results:

    • Head-mounted sensors achieved high accuracy (98.91% pre-turn, 98.73% during-turn).
    • Sliding-window LSTM on the right heel reached 96.93% accuracy and 97.05% precision.
    • Sliding-window GRU on the left toe achieved 95.85% accuracy and 95.89% precision.

    Conclusions:

    • Optimized sliding-window sizes effectively predict pre-head-turning kinematics using wearable IMU data.
    • Head and chest sensors are critical for recognizing pre-head-turning kinematics and direction, showing promise for gait applications.
    • The system demonstrates robust performance across various head-turning directions, suitable for gait-adaptive systems.