Distinct feature extraction for video-based gait phase classification

Minxiang Ye, Cheng Yang, Vladimir Stankovic, Lina Stankovic, Samuel Cheng

Research output: Contribution to journalArticle

Abstract

Recent advances in image acquisition and analysis have resulted in disruptive innovation in physical rehabilitation systems facilitating cost-effective, portable, video-based gait assessment. While these inexpensive motion capture systems, suitable for home rehabilitation, do not generally provide accurate kinematics measurements on their own, image processing algorithms ensure gait analysis that is accurate enough for rehabilitation programs. This paper proposes high-accuracy classification of gait phases and muscle actions, using readings from low-cost motion capture systems. First, 12 gait parameters, drawn from the medical literature, are defined to characterize gait patterns. These proposed parameters are then used as input to our proposed multi-channel time-series classification and gait phase reconstruction methods. Proposed methods fully utilize temporal information of gait parameters, thus improving the final classification accuracy. The validation, conducted using 126 experiments, with 6 healthy volunteers and 9 stroke survivors with manually-labelled gait phases, achieves state-of-art classification accuracy of gait phase with lower computational complexity compared to previous solutions
LanguageEnglish
Number of pages13
JournalIEEE Transactions on Multimedia
Early online date19 Sep 2019
DOIs
Publication statusE-pub ahead of print - 19 Sep 2019

Fingerprint

Feature extraction
Patient rehabilitation
Gait analysis
Image acquisition
Image analysis
Muscle
Costs
Time series
Computational complexity
Kinematics
Image processing
Innovation
Experiments

Keywords

  • feature extraction
  • gait phase classification
  • motion capture system

Cite this

@article{54719e754aab4de59d49307283cd202e,
title = "Distinct feature extraction for video-based gait phase classification",
abstract = "Recent advances in image acquisition and analysis have resulted in disruptive innovation in physical rehabilitation systems facilitating cost-effective, portable, video-based gait assessment. While these inexpensive motion capture systems, suitable for home rehabilitation, do not generally provide accurate kinematics measurements on their own, image processing algorithms ensure gait analysis that is accurate enough for rehabilitation programs. This paper proposes high-accuracy classification of gait phases and muscle actions, using readings from low-cost motion capture systems. First, 12 gait parameters, drawn from the medical literature, are defined to characterize gait patterns. These proposed parameters are then used as input to our proposed multi-channel time-series classification and gait phase reconstruction methods. Proposed methods fully utilize temporal information of gait parameters, thus improving the final classification accuracy. The validation, conducted using 126 experiments, with 6 healthy volunteers and 9 stroke survivors with manually-labelled gait phases, achieves state-of-art classification accuracy of gait phase with lower computational complexity compared to previous solutions",
keywords = "feature extraction, gait phase classification, motion capture system",
author = "Minxiang Ye and Cheng Yang and Vladimir Stankovic and Lina Stankovic and Samuel Cheng",
note = "{\circledC} 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.",
year = "2019",
month = "9",
day = "19",
doi = "10.1109/TMM.2019.2942479",
language = "English",
journal = "IEEE Transactions on Multimedia",
issn = "1520-9210",

}

Distinct feature extraction for video-based gait phase classification. / Ye, Minxiang; Yang, Cheng; Stankovic, Vladimir; Stankovic, Lina; Cheng, Samuel.

In: IEEE Transactions on Multimedia, 19.09.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Distinct feature extraction for video-based gait phase classification

AU - Ye, Minxiang

AU - Yang, Cheng

AU - Stankovic, Vladimir

AU - Stankovic, Lina

AU - Cheng, Samuel

N1 - © 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

PY - 2019/9/19

Y1 - 2019/9/19

N2 - Recent advances in image acquisition and analysis have resulted in disruptive innovation in physical rehabilitation systems facilitating cost-effective, portable, video-based gait assessment. While these inexpensive motion capture systems, suitable for home rehabilitation, do not generally provide accurate kinematics measurements on their own, image processing algorithms ensure gait analysis that is accurate enough for rehabilitation programs. This paper proposes high-accuracy classification of gait phases and muscle actions, using readings from low-cost motion capture systems. First, 12 gait parameters, drawn from the medical literature, are defined to characterize gait patterns. These proposed parameters are then used as input to our proposed multi-channel time-series classification and gait phase reconstruction methods. Proposed methods fully utilize temporal information of gait parameters, thus improving the final classification accuracy. The validation, conducted using 126 experiments, with 6 healthy volunteers and 9 stroke survivors with manually-labelled gait phases, achieves state-of-art classification accuracy of gait phase with lower computational complexity compared to previous solutions

AB - Recent advances in image acquisition and analysis have resulted in disruptive innovation in physical rehabilitation systems facilitating cost-effective, portable, video-based gait assessment. While these inexpensive motion capture systems, suitable for home rehabilitation, do not generally provide accurate kinematics measurements on their own, image processing algorithms ensure gait analysis that is accurate enough for rehabilitation programs. This paper proposes high-accuracy classification of gait phases and muscle actions, using readings from low-cost motion capture systems. First, 12 gait parameters, drawn from the medical literature, are defined to characterize gait patterns. These proposed parameters are then used as input to our proposed multi-channel time-series classification and gait phase reconstruction methods. Proposed methods fully utilize temporal information of gait parameters, thus improving the final classification accuracy. The validation, conducted using 126 experiments, with 6 healthy volunteers and 9 stroke survivors with manually-labelled gait phases, achieves state-of-art classification accuracy of gait phase with lower computational complexity compared to previous solutions

KW - feature extraction

KW - gait phase classification

KW - motion capture system

UR - https://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=6046

U2 - 10.1109/TMM.2019.2942479

DO - 10.1109/TMM.2019.2942479

M3 - Article

JO - IEEE Transactions on Multimedia

T2 - IEEE Transactions on Multimedia

JF - IEEE Transactions on Multimedia

SN - 1520-9210

ER -