Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Apr 17, 2026

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

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

Published on: April 3, 2026

Deep learning-based gait phase detection using shank-mounted IMU data: Classification approach.

Wonseok Choi¹, Mun-Taek Choi¹

¹Department of Intelligent Robotics, Sungkyunkwan University, Suwon, South Korea.

|April 15, 2026

Summary

This summary is machine-generated.

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

Motion Planning-Augmented Hierarchical Reinforcement Learning for Long-Horizon Mobile Manipulation.

Sensors (Basel, Switzerland)·2026

Same author

The effectiveness of school-based physical activity interventions on health-related fitness and BMI in South Korea: a three-level meta-analysis.

Frontiers in pediatrics·2026

Same author

Ultrasonic modulation of brain glymphatic transport: from observations to theranostic applications.

Cancer drug resistance (Alhambra, Calif.)·2026

Same author

Volumetric photoacoustic/ultrasound image fusion and enhancement with dual-tree complex wavelet transform.

Biomedical optics express·2026

Same author

The genitals of rhinoceros beetles: a general overview of the endophallus in the tribe Agaocephalini (Coleoptera: Scarabaeidae: Dynastinae).

PeerJ·2026

Same author

Clinical effect of bisphosphonate in patients with hormone receptor-positive early breast cancer: A nationwide real-world data analysis.

Breast (Edinburgh, Scotland)·2026

Same journal

Thymidylate synthase inhibitory drugs induce p53-dependent pathways differently.

PloS one·2026

Same journal

Top-down and bottom-up attention for joint pattern classification and reconstruction.

PloS one·2026

Same journal

Short- and long-term scaling behavior of blood pressure and pulse arrival time during sleep in healthy controls and patients with obstructive sleep apnea.

PloS one·2026

Same journal

Double DQN-based secrecy energy efficiency and fairness performance in IRS-assisted NOMA systems with friendly jamming.

PloS one·2026

Same journal

10 recommendations for strengthening citizen science for improved societal and ecological outcomes: A co-produced analysis of challenges and opportunities in the 21st century.

PloS one·2026

Same journal

Paying in public: Peer effects, impression management, and willingness to pay on digital payment platforms.

PloS one·2026

See all related articles

Accurate gait phase detection is crucial for mobility assessment. This study shows that an end-to-end Transformer model using shank-mounted IMUs achieves high accuracy, simplifying analysis and improving real-world applicability for gait analysis.

Area of Science:

Biomechanics and Movement Science
Wearable Technology and Sensor Systems
Machine Learning and Artificial Intelligence

Background:

Gait analysis is vital for assessing mobility and health, requiring accurate gait cycle phase detection.
Current methods often involve complex multi-sensor setups or lack robustness and real-time performance.
Existing machine learning and CNN approaches struggle with temporal dependencies in gait data.

Purpose of the Study:

To investigate an end-to-end supervised classification learning approach for gait phase detection.
To evaluate the performance of a Transformer model against CNN and Hybrid LSTM+GRU models using shank-mounted IMU data.
To demonstrate a simplified and robust method for gait phase recognition.

Main Methods:

More Related Videos

3D Kinematic Gait Analysis for Preclinical Studies in Rodents

3D Kinematic Gait Analysis for Preclinical Studies in Rodents

Published on: August 3, 2019

Automated Gait Analysis in Mice with Chronic Constriction Injury

Automated Gait Analysis in Mice with Chronic Constriction Injury

Published on: October 17, 2017

Related Experiment Videos

Last Updated: Apr 17, 2026

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

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

Published on: April 3, 2026

3D Kinematic Gait Analysis for Preclinical Studies in Rodents

3D Kinematic Gait Analysis for Preclinical Studies in Rodents

Published on: August 3, 2019

Automated Gait Analysis in Mice with Chronic Constriction Injury

Automated Gait Analysis in Mice with Chronic Constriction Injury

Published on: October 17, 2017

Utilized the NONAN GaitPrint dataset with time-series gait recordings from 35 healthy adults.

Employed an end-to-end supervised classification learning strategy with a Transformer model.

Trained and compared 1D CNN, Hybrid LSTM+GRU, and Transformer models using shank-mounted acceleration and angular velocity signals.

Main Results:

The Transformer model achieved an F1-score of approximately 92.99% for gait phase detection.
CNN and Hybrid LSTM+GRU models showed comparable performance to the Transformer.
No substantial performance differences were observed among the evaluated deep learning models.

Conclusions:

End-to-end learning with shank-mounted IMUs enables clinically and practically feasible gait phase detection.
The Transformer model offers a robust and simplified approach to gait analysis.
This method enhances real-world applicability by reducing system complexity.