A vision-based regression model to evaluate Parkinsonian gait from monocular image sequences

You Yin Chen, Chien Wen Cho, Sheng Huang Lin, Hsin Yi Lai, Yu Chun Lo, Shin Yuan Chen, Yuan-Jen Chang, Wen Tzeng Huang, Chin-Hsing Chen, Fu Shan Jaw, Siny Tsang, Sheng Tsung Tsai

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

Parkinson's Disease (PD) is a common neurodegenerative disorder with progressive loss of dopaminergic and other sub-cortical neurons. Among various approaches, gait analysis is commonly used to help identify the biometric features of PD. There have been some studies to date on both the classification of PD and estimation of gait parameters. However, it is also important to construct a regression system that can evaluate the degree of abnormality in PD patients. In this paper, we intended to develop a PD gait regression model that is capable of predicting the severity of motor dysfunction from given gait image sequences. We used a model-free strategy and thus avoided the critical demands of segmentation and parameter estimation. Furthermore, we used linear discriminant analysis (LDA) to increase the feature efficiency by maximizing and minimizing the between- and within-group variations. Regression was also achieved by assessing the spatial and temporal information through classification and finally by using these two new indices for linear regression. According to the experiments, the outcomes significantly correlated with the sum of sub-scores from the Unified Parkinson's Disease Rating Scale (UPDRS): motor examination section with r = 0.92 and 0.85 for training and testing, respectively, with p < 0.0001. Compared with conventional methods, our system provided a better evaluation of PD abnormality.

Original languageEnglish
Pages (from-to)520-526
Number of pages7
JournalExpert Systems with Applications
Volume39
Issue number1
DOIs
Publication statusPublished - Jan 2012
Externally publishedYes

Fingerprint

Gait analysis
Discriminant analysis
Biometrics
Linear regression
Parameter estimation
Neurons
Testing
Experiments

Keywords

  • Classification
  • Human motion analysis
  • Linear discriminant analysis (LDA)
  • Parkinsonian gait
  • Regression

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Engineering(all)

Cite this

A vision-based regression model to evaluate Parkinsonian gait from monocular image sequences. / Chen, You Yin; Cho, Chien Wen; Lin, Sheng Huang; Lai, Hsin Yi; Lo, Yu Chun; Chen, Shin Yuan; Chang, Yuan-Jen; Huang, Wen Tzeng; Chen, Chin-Hsing; Jaw, Fu Shan; Tsang, Siny; Tsai, Sheng Tsung.

In: Expert Systems with Applications, Vol. 39, No. 1, 01.2012, p. 520-526.

Research output: Contribution to journalArticle

Chen, YY, Cho, CW, Lin, SH, Lai, HY, Lo, YC, Chen, SY, Chang, Y-J, Huang, WT, Chen, C-H, Jaw, FS, Tsang, S & Tsai, ST 2012, 'A vision-based regression model to evaluate Parkinsonian gait from monocular image sequences', Expert Systems with Applications, vol. 39, no. 1, pp. 520-526. https://doi.org/10.1016/j.eswa.2011.07.042
Chen, You Yin ; Cho, Chien Wen ; Lin, Sheng Huang ; Lai, Hsin Yi ; Lo, Yu Chun ; Chen, Shin Yuan ; Chang, Yuan-Jen ; Huang, Wen Tzeng ; Chen, Chin-Hsing ; Jaw, Fu Shan ; Tsang, Siny ; Tsai, Sheng Tsung. / A vision-based regression model to evaluate Parkinsonian gait from monocular image sequences. In: Expert Systems with Applications. 2012 ; Vol. 39, No. 1. pp. 520-526.
@article{d8b57075758346e6b81b46deb05a793f,
title = "A vision-based regression model to evaluate Parkinsonian gait from monocular image sequences",
abstract = "Parkinson's Disease (PD) is a common neurodegenerative disorder with progressive loss of dopaminergic and other sub-cortical neurons. Among various approaches, gait analysis is commonly used to help identify the biometric features of PD. There have been some studies to date on both the classification of PD and estimation of gait parameters. However, it is also important to construct a regression system that can evaluate the degree of abnormality in PD patients. In this paper, we intended to develop a PD gait regression model that is capable of predicting the severity of motor dysfunction from given gait image sequences. We used a model-free strategy and thus avoided the critical demands of segmentation and parameter estimation. Furthermore, we used linear discriminant analysis (LDA) to increase the feature efficiency by maximizing and minimizing the between- and within-group variations. Regression was also achieved by assessing the spatial and temporal information through classification and finally by using these two new indices for linear regression. According to the experiments, the outcomes significantly correlated with the sum of sub-scores from the Unified Parkinson's Disease Rating Scale (UPDRS): motor examination section with r = 0.92 and 0.85 for training and testing, respectively, with p < 0.0001. Compared with conventional methods, our system provided a better evaluation of PD abnormality.",
keywords = "Classification, Human motion analysis, Linear discriminant analysis (LDA), Parkinsonian gait, Regression",
author = "Chen, {You Yin} and Cho, {Chien Wen} and Lin, {Sheng Huang} and Lai, {Hsin Yi} and Lo, {Yu Chun} and Chen, {Shin Yuan} and Yuan-Jen Chang and Huang, {Wen Tzeng} and Chin-Hsing Chen and Jaw, {Fu Shan} and Siny Tsang and Tsai, {Sheng Tsung}",
year = "2012",
month = "1",
doi = "10.1016/j.eswa.2011.07.042",
language = "English",
volume = "39",
pages = "520--526",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Limited",
number = "1",

}

TY - JOUR

T1 - A vision-based regression model to evaluate Parkinsonian gait from monocular image sequences

AU - Chen, You Yin

AU - Cho, Chien Wen

AU - Lin, Sheng Huang

AU - Lai, Hsin Yi

AU - Lo, Yu Chun

AU - Chen, Shin Yuan

AU - Chang, Yuan-Jen

AU - Huang, Wen Tzeng

AU - Chen, Chin-Hsing

AU - Jaw, Fu Shan

AU - Tsang, Siny

AU - Tsai, Sheng Tsung

PY - 2012/1

Y1 - 2012/1

N2 - Parkinson's Disease (PD) is a common neurodegenerative disorder with progressive loss of dopaminergic and other sub-cortical neurons. Among various approaches, gait analysis is commonly used to help identify the biometric features of PD. There have been some studies to date on both the classification of PD and estimation of gait parameters. However, it is also important to construct a regression system that can evaluate the degree of abnormality in PD patients. In this paper, we intended to develop a PD gait regression model that is capable of predicting the severity of motor dysfunction from given gait image sequences. We used a model-free strategy and thus avoided the critical demands of segmentation and parameter estimation. Furthermore, we used linear discriminant analysis (LDA) to increase the feature efficiency by maximizing and minimizing the between- and within-group variations. Regression was also achieved by assessing the spatial and temporal information through classification and finally by using these two new indices for linear regression. According to the experiments, the outcomes significantly correlated with the sum of sub-scores from the Unified Parkinson's Disease Rating Scale (UPDRS): motor examination section with r = 0.92 and 0.85 for training and testing, respectively, with p < 0.0001. Compared with conventional methods, our system provided a better evaluation of PD abnormality.

AB - Parkinson's Disease (PD) is a common neurodegenerative disorder with progressive loss of dopaminergic and other sub-cortical neurons. Among various approaches, gait analysis is commonly used to help identify the biometric features of PD. There have been some studies to date on both the classification of PD and estimation of gait parameters. However, it is also important to construct a regression system that can evaluate the degree of abnormality in PD patients. In this paper, we intended to develop a PD gait regression model that is capable of predicting the severity of motor dysfunction from given gait image sequences. We used a model-free strategy and thus avoided the critical demands of segmentation and parameter estimation. Furthermore, we used linear discriminant analysis (LDA) to increase the feature efficiency by maximizing and minimizing the between- and within-group variations. Regression was also achieved by assessing the spatial and temporal information through classification and finally by using these two new indices for linear regression. According to the experiments, the outcomes significantly correlated with the sum of sub-scores from the Unified Parkinson's Disease Rating Scale (UPDRS): motor examination section with r = 0.92 and 0.85 for training and testing, respectively, with p < 0.0001. Compared with conventional methods, our system provided a better evaluation of PD abnormality.

KW - Classification

KW - Human motion analysis

KW - Linear discriminant analysis (LDA)

KW - Parkinsonian gait

KW - Regression

UR - http://www.scopus.com/inward/record.url?scp=81855221659&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=81855221659&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2011.07.042

DO - 10.1016/j.eswa.2011.07.042

M3 - Article

AN - SCOPUS:81855221659

VL - 39

SP - 520

EP - 526

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

IS - 1

ER -