Reliability and Validity Examination of a New Gait Motion Analysis System

Recent advancements have made two-dimensional (2D) clinical gait analysis systems more accessible and portable than traditional three-dimensional (3D) clinical systems. This study evaluates the reliability and validity of gait measurements using monocular and composite camera setups with VisionPose,...

Full description

Saved in:
Bibliographic Details
Main Authors: Tadamitsu Matsuda, Yuji Fujino, Tomoyuki Morisawa, Tetsuya Takahashi, Kei Kakegawa, Takanari Matsumoto, Takehiko Kiyohara, Hiroshi Fukushima, Makoto Higuchi, Yasuo Torimoto, Masaki Miwa, Toshiyuki Fujiwara, Hiroyuki Daida
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/4/1076
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850080946947096576
author Tadamitsu Matsuda
Yuji Fujino
Tomoyuki Morisawa
Tetsuya Takahashi
Kei Kakegawa
Takanari Matsumoto
Takehiko Kiyohara
Hiroshi Fukushima
Makoto Higuchi
Yasuo Torimoto
Masaki Miwa
Toshiyuki Fujiwara
Hiroyuki Daida
author_facet Tadamitsu Matsuda
Yuji Fujino
Tomoyuki Morisawa
Tetsuya Takahashi
Kei Kakegawa
Takanari Matsumoto
Takehiko Kiyohara
Hiroshi Fukushima
Makoto Higuchi
Yasuo Torimoto
Masaki Miwa
Toshiyuki Fujiwara
Hiroyuki Daida
author_sort Tadamitsu Matsuda
collection DOAJ
description Recent advancements have made two-dimensional (2D) clinical gait analysis systems more accessible and portable than traditional three-dimensional (3D) clinical systems. This study evaluates the reliability and validity of gait measurements using monocular and composite camera setups with VisionPose, comparing them to the Vicon 3D motion capture system as a reference. Key gait parameters—including hip and knee joint angles, and time and distance factors—were assessed under normal, maximum speed, and tandem gait conditions during level walking. The results show that the intraclass correlation coefficient (ICC(1,k)) for the 2D model exceeded 0.969 for the monocular camera and 0.963 for the composite camera for gait parameters. Time–distance gait parameters demonstrated excellent relative agreement across walking styles, while joint range of motion showed overall strong agreement. However, accuracy was lower for measurements during tandem walking. The Cronbach’s alpha coefficient for time–distance parameters ranged from 0.932 to 0.999 (monocular) and from 0.823 to 0.998 (composite). In contrast, for joint range of motion, the coefficient varied more widely, ranging from 0.826 to 0.985 (monocular) and from 0.314 to 0.974 (composite). The correlation coefficients for spatiotemporal gait parameters were greater than 0.933 (monocular) and 0.837 (composite). However, for joint angle parameters, the coefficients were lower during tandem walking. This study underscores the potential of 2D models in clinical applications and highlights areas for improvement to enhance their reliability and application scope.
format Article
id doaj-art-578e8805d21a4381b51ab1a95e2c7f07
institution DOAJ
issn 1424-8220
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-578e8805d21a4381b51ab1a95e2c7f072025-08-20T02:44:50ZengMDPI AGSensors1424-82202025-02-01254107610.3390/s25041076Reliability and Validity Examination of a New Gait Motion Analysis SystemTadamitsu Matsuda0Yuji Fujino1Tomoyuki Morisawa2Tetsuya Takahashi3Kei Kakegawa4Takanari Matsumoto5Takehiko Kiyohara6Hiroshi Fukushima7Makoto Higuchi8Yasuo Torimoto9Masaki Miwa10Toshiyuki Fujiwara11Hiroyuki Daida12Department of Physical Therapy, Faculty of Health Science, Juntendo University, Tokyo 113-8421, JapanDepartment of Physical Therapy, Faculty of Health Science, Juntendo University, Tokyo 113-8421, JapanDepartment of Physical Therapy, Faculty of Health Science, Juntendo University, Tokyo 113-8421, JapanDepartment of Physical Therapy, Faculty of Health Science, Juntendo University, Tokyo 113-8421, JapanDepartment of Physical Therapy, Faculty of Health Science, Juntendo University, Tokyo 113-8421, JapanGlobal Development Center, Development Department, Development Section, IMASEN Electric Industrial Co., Ltd., Inuyama 484-0083, JapanGlobal Development Center, Development Department, Development Section, IMASEN Electric Industrial Co., Ltd., Inuyama 484-0083, JapanGlobal Development Center, Development Department, Development Section, IMASEN Electric Industrial Co., Ltd., Inuyama 484-0083, JapanGlobal Development Center, Development Department, Development Section, IMASEN Electric Industrial Co., Ltd., Inuyama 484-0083, JapanGlobal Development Center, Development Department, Development Section, IMASEN Electric Industrial Co., Ltd., Inuyama 484-0083, JapanGlobal Development Center, Development Department, Development Section, IMASEN Electric Industrial Co., Ltd., Inuyama 484-0083, JapanDepartment of Rehabilitation Medicine, Graduate School of Medicine, Juntendo University, Tokyo 113-8421, JapanDepartment of Physical Therapy, Faculty of Health Science, Juntendo University, Tokyo 113-8421, JapanRecent advancements have made two-dimensional (2D) clinical gait analysis systems more accessible and portable than traditional three-dimensional (3D) clinical systems. This study evaluates the reliability and validity of gait measurements using monocular and composite camera setups with VisionPose, comparing them to the Vicon 3D motion capture system as a reference. Key gait parameters—including hip and knee joint angles, and time and distance factors—were assessed under normal, maximum speed, and tandem gait conditions during level walking. The results show that the intraclass correlation coefficient (ICC(1,k)) for the 2D model exceeded 0.969 for the monocular camera and 0.963 for the composite camera for gait parameters. Time–distance gait parameters demonstrated excellent relative agreement across walking styles, while joint range of motion showed overall strong agreement. However, accuracy was lower for measurements during tandem walking. The Cronbach’s alpha coefficient for time–distance parameters ranged from 0.932 to 0.999 (monocular) and from 0.823 to 0.998 (composite). In contrast, for joint range of motion, the coefficient varied more widely, ranging from 0.826 to 0.985 (monocular) and from 0.314 to 0.974 (composite). The correlation coefficients for spatiotemporal gait parameters were greater than 0.933 (monocular) and 0.837 (composite). However, for joint angle parameters, the coefficients were lower during tandem walking. This study underscores the potential of 2D models in clinical applications and highlights areas for improvement to enhance their reliability and application scope.https://www.mdpi.com/1424-8220/25/4/1076motion analysisgaitvalidationhuman pose estimation
spellingShingle Tadamitsu Matsuda
Yuji Fujino
Tomoyuki Morisawa
Tetsuya Takahashi
Kei Kakegawa
Takanari Matsumoto
Takehiko Kiyohara
Hiroshi Fukushima
Makoto Higuchi
Yasuo Torimoto
Masaki Miwa
Toshiyuki Fujiwara
Hiroyuki Daida
Reliability and Validity Examination of a New Gait Motion Analysis System
Sensors
motion analysis
gait
validation
human pose estimation
title Reliability and Validity Examination of a New Gait Motion Analysis System
title_full Reliability and Validity Examination of a New Gait Motion Analysis System
title_fullStr Reliability and Validity Examination of a New Gait Motion Analysis System
title_full_unstemmed Reliability and Validity Examination of a New Gait Motion Analysis System
title_short Reliability and Validity Examination of a New Gait Motion Analysis System
title_sort reliability and validity examination of a new gait motion analysis system
topic motion analysis
gait
validation
human pose estimation
url https://www.mdpi.com/1424-8220/25/4/1076
work_keys_str_mv AT tadamitsumatsuda reliabilityandvalidityexaminationofanewgaitmotionanalysissystem
AT yujifujino reliabilityandvalidityexaminationofanewgaitmotionanalysissystem
AT tomoyukimorisawa reliabilityandvalidityexaminationofanewgaitmotionanalysissystem
AT tetsuyatakahashi reliabilityandvalidityexaminationofanewgaitmotionanalysissystem
AT keikakegawa reliabilityandvalidityexaminationofanewgaitmotionanalysissystem
AT takanarimatsumoto reliabilityandvalidityexaminationofanewgaitmotionanalysissystem
AT takehikokiyohara reliabilityandvalidityexaminationofanewgaitmotionanalysissystem
AT hiroshifukushima reliabilityandvalidityexaminationofanewgaitmotionanalysissystem
AT makotohiguchi reliabilityandvalidityexaminationofanewgaitmotionanalysissystem
AT yasuotorimoto reliabilityandvalidityexaminationofanewgaitmotionanalysissystem
AT masakimiwa reliabilityandvalidityexaminationofanewgaitmotionanalysissystem
AT toshiyukifujiwara reliabilityandvalidityexaminationofanewgaitmotionanalysissystem
AT hiroyukidaida reliabilityandvalidityexaminationofanewgaitmotionanalysissystem