Single-view, video-based diagnosis of Parkinson's Disease based on arm and leg joint tracking

Jun Seok Seo; Yiyu Chen; Do Young Kwon; Christian Wallraven

doi:10.1109/CMAEE58250.2022.00037

Single-view, video-based diagnosis of Parkinson's Disease based on arm and leg joint tracking

Jun Seok Seo^*
, Yiyu Chen^*
, Do Young Kwon
, Christian Wallraven^*

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

3 Citations (Scopus)

Abstract

Automatic diagnosis of Parkinson's Disease (PD) from sensor data is an important topic given the growing numbers of patients, and the increasing costs to the quality of life of an aging society. Several approaches have been proposed aimed at such an automatic diagnosis, but often suffer from complicated sensor setups or setups ill-fitting for the limitations in clinical settings. Here, we present a system that uses frequency-based analysis of joint data from both arms and legs from a single, frontally-viewed video of people walking towards a camera. We evaluate three machine-learning models on frequency-based features extracted from the joint dynamics on two larger datasets containing a total of N=300 videos of over 50 PD patients and healthy control people. Results confirm typical clinical expectations (leg frequencies are slower in PD patients) and in addition show excellent generalizability even across datasets with performance of up to 97% for an Ensemble classifier.

Original language	English
Title of host publication	Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	172-176
Number of pages	5
ISBN (Electronic)	9798350346572
DOIs	https://doi.org/10.1109/CMAEE58250.2022.00037
Publication status	Published - 2022
Event	2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022 - Virtual, Online, China Duration: 2022 Dec 16 → 2022 Dec 18

Publication series

Name	Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022

Conference

Conference	2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022
Country/Territory	China
City	Virtual, Online
Period	22/12/16 → 22/12/18

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

Keywords

2d
Parkinson's
Parkinson's Disease
arm swing
artificial intelligence
gait
machine learning
openpose
pose analysis
single camera
video analysis
walk

ASJC Scopus subject areas

Computer Science Applications
Electrical and Electronic Engineering
Mechanical Engineering
Media Technology
Control and Optimization

Access to Document

10.1109/CMAEE58250.2022.00037

Cite this

Seo, J. S., Chen, Y., Kwon, D. Y., & Wallraven, C. (2022). Single-view, video-based diagnosis of Parkinson's Disease based on arm and leg joint tracking. In Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022 (pp. 172-176). (Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CMAEE58250.2022.00037

Single-view, video-based diagnosis of Parkinson's Disease based on arm and leg joint tracking. / Seo, Jun Seok; Chen, Yiyu; Kwon, Do Young et al.
Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022. Institute of Electrical and Electronics Engineers Inc., 2022. p. 172-176 (Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Seo, JS, Chen, Y, Kwon, DY & Wallraven, C 2022, Single-view, video-based diagnosis of Parkinson's Disease based on arm and leg joint tracking. in Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022. Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022, Institute of Electrical and Electronics Engineers Inc., pp. 172-176, 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022, Virtual, Online, China, 22/12/16. https://doi.org/10.1109/CMAEE58250.2022.00037

Seo JS, Chen Y, Kwon DY, Wallraven C. Single-view, video-based diagnosis of Parkinson's Disease based on arm and leg joint tracking. In Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022. Institute of Electrical and Electronics Engineers Inc. 2022. p. 172-176. (Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022). doi: 10.1109/CMAEE58250.2022.00037

Seo, Jun Seok ; Chen, Yiyu ; Kwon, Do Young et al. / Single-view, video-based diagnosis of Parkinson's Disease based on arm and leg joint tracking. Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 172-176 (Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022).

@inproceedings{c2802851a073443fa0e6d5725b90fff2,

title = "Single-view, video-based diagnosis of Parkinson's Disease based on arm and leg joint tracking",

abstract = "Automatic diagnosis of Parkinson's Disease (PD) from sensor data is an important topic given the growing numbers of patients, and the increasing costs to the quality of life of an aging society. Several approaches have been proposed aimed at such an automatic diagnosis, but often suffer from complicated sensor setups or setups ill-fitting for the limitations in clinical settings. Here, we present a system that uses frequency-based analysis of joint data from both arms and legs from a single, frontally-viewed video of people walking towards a camera. We evaluate three machine-learning models on frequency-based features extracted from the joint dynamics on two larger datasets containing a total of N=300 videos of over 50 PD patients and healthy control people. Results confirm typical clinical expectations (leg frequencies are slower in PD patients) and in addition show excellent generalizability even across datasets with performance of up to 97\% for an Ensemble classifier.",

keywords = "2d, Parkinson's, Parkinson's Disease, arm swing, artificial intelligence, gait, machine learning, openpose, pose analysis, single camera, video analysis, walk",

author = "Seo, \{Jun Seok\} and Yiyu Chen and Kwon, \{Do Young\} and Christian Wallraven",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022 ; Conference date: 16-12-2022 Through 18-12-2022",

year = "2022",

doi = "10.1109/CMAEE58250.2022.00037",

language = "English",

series = "Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "172--176",

booktitle = "Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022",

}

TY - GEN

T1 - Single-view, video-based diagnosis of Parkinson's Disease based on arm and leg joint tracking

AU - Seo, Jun Seok

AU - Chen, Yiyu

AU - Kwon, Do Young

AU - Wallraven, Christian

PY - 2022

Y1 - 2022

N2 - Automatic diagnosis of Parkinson's Disease (PD) from sensor data is an important topic given the growing numbers of patients, and the increasing costs to the quality of life of an aging society. Several approaches have been proposed aimed at such an automatic diagnosis, but often suffer from complicated sensor setups or setups ill-fitting for the limitations in clinical settings. Here, we present a system that uses frequency-based analysis of joint data from both arms and legs from a single, frontally-viewed video of people walking towards a camera. We evaluate three machine-learning models on frequency-based features extracted from the joint dynamics on two larger datasets containing a total of N=300 videos of over 50 PD patients and healthy control people. Results confirm typical clinical expectations (leg frequencies are slower in PD patients) and in addition show excellent generalizability even across datasets with performance of up to 97% for an Ensemble classifier.

AB - Automatic diagnosis of Parkinson's Disease (PD) from sensor data is an important topic given the growing numbers of patients, and the increasing costs to the quality of life of an aging society. Several approaches have been proposed aimed at such an automatic diagnosis, but often suffer from complicated sensor setups or setups ill-fitting for the limitations in clinical settings. Here, we present a system that uses frequency-based analysis of joint data from both arms and legs from a single, frontally-viewed video of people walking towards a camera. We evaluate three machine-learning models on frequency-based features extracted from the joint dynamics on two larger datasets containing a total of N=300 videos of over 50 PD patients and healthy control people. Results confirm typical clinical expectations (leg frequencies are slower in PD patients) and in addition show excellent generalizability even across datasets with performance of up to 97% for an Ensemble classifier.

KW - 2d

KW - Parkinson's

KW - Parkinson's Disease

KW - arm swing

KW - artificial intelligence

KW - gait

KW - machine learning

KW - openpose

KW - pose analysis

KW - single camera

KW - video analysis

KW - walk

UR - https://www.scopus.com/pages/publications/85151390122

U2 - 10.1109/CMAEE58250.2022.00037

DO - 10.1109/CMAEE58250.2022.00037

M3 - Conference contribution

AN - SCOPUS:85151390122

T3 - Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022

SP - 172

EP - 176

BT - Proceedings - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2022 International Conference on Mechanical, Automation and Electrical Engineering, CMAEE 2022

Y2 - 16 December 2022 through 18 December 2022

ER -

Single-view, video-based diagnosis of Parkinson's Disease based on arm and leg joint tracking

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this