TY - JOUR
T1 - Feature visualization and classification for the discrimination between individuals with Parkinson's disease under levodopa and DBS treatments
AU - Machado, Alessandro R.P.
AU - Zaidan, Hudson Capanema
AU - Paix�o, Ana Paula Souza
AU - Cavalheiro, Guilherme Lopes
AU - Oliveira, F�bio Henrique Monteiro
AU - J�nior, Jo�o Areis Ferreira Barbosa
AU - Naves, Kheline
AU - Pereira, Adriano Alves
AU - Pereira, Janser Moura
AU - Pouratian, Nader
AU - Zhuo, Xiaoyi
AU - O'Keeffe, Andrew
AU - Sharim, Justin
AU - Bordelon, Yvette
AU - Yang, Laurice
AU - Vieira, Marcus Fraga
AU - Andrade, Adriano O.
N1 - Funding Information:
The authors are grateful to Brazilian Government Agencies CAPES, CNPq, FAPEMIG and FAPEG for supporting the study. The authors also extend appreciation to the subjects who volunteered for this study. NP is supported in part by a NIBIB award (K23EB014326) as well as philanthropic support from the Casa Colina Centers for Rehabilitation.
Publisher Copyright:
� 2016 The Author(s).
PY - 2016/12/30
Y1 - 2016/12/30
N2 - Background: Over the years, a number of distinct treatments have been adopted for the management of the motor symptoms of Parkinson's disease (PD), including pharmacologic therapies and deep brain stimulation (DBS). Efficacy is most often evaluated by subjective assessments, which are prone to error and dependent on the experience of the examiner. Our goal was to identify an objective means of assessing response to therapy. Methods: In this study, we employed objective analyses in order to visualize and identify differences between three groups: healthy control (N=10), subjects with PD treated with DBS (N=12), and subjects with PD treated with levodopa (N=16). Subjects were assessed during execution of three dynamic tasks (finger taps, finger to nose, supination and pronation) and a static task (extended arm with no active movement). Measurements were acquired with two pairs of inertial and electromyographic sensors. Feature extraction was applied to estimate the relevant information from the data after which the high-dimensional feature space was reduced to a two-dimensional space using the nonlinear Sammon's map. Non-parametric analysis of variance was employed for the verification of relevant statistical differences among the groups (p<0.05). In addition, K-fold cross-validation for discriminant analysis based on Gaussian Finite Mixture Modeling was employed for data classification. Results: The results showed visual and statistical differences for all groups and conditions (i.e., static and dynamic tasks). The employed methods were successful for the discrimination of the groups. Classification accuracy was 81�6% (mean�standard deviation) and 71�8%, for training and test groups respectively. Conclusions: This research showed the discrimination between healthy and diseased groups conditions. The methods were also able to discriminate individuals with PD treated with DBS and levodopa. These methods enable objective characterization and visualization of features extracted from inertial and electromyographic sensors for different groups.
AB - Background: Over the years, a number of distinct treatments have been adopted for the management of the motor symptoms of Parkinson's disease (PD), including pharmacologic therapies and deep brain stimulation (DBS). Efficacy is most often evaluated by subjective assessments, which are prone to error and dependent on the experience of the examiner. Our goal was to identify an objective means of assessing response to therapy. Methods: In this study, we employed objective analyses in order to visualize and identify differences between three groups: healthy control (N=10), subjects with PD treated with DBS (N=12), and subjects with PD treated with levodopa (N=16). Subjects were assessed during execution of three dynamic tasks (finger taps, finger to nose, supination and pronation) and a static task (extended arm with no active movement). Measurements were acquired with two pairs of inertial and electromyographic sensors. Feature extraction was applied to estimate the relevant information from the data after which the high-dimensional feature space was reduced to a two-dimensional space using the nonlinear Sammon's map. Non-parametric analysis of variance was employed for the verification of relevant statistical differences among the groups (p<0.05). In addition, K-fold cross-validation for discriminant analysis based on Gaussian Finite Mixture Modeling was employed for data classification. Results: The results showed visual and statistical differences for all groups and conditions (i.e., static and dynamic tasks). The employed methods were successful for the discrimination of the groups. Classification accuracy was 81�6% (mean�standard deviation) and 71�8%, for training and test groups respectively. Conclusions: This research showed the discrimination between healthy and diseased groups conditions. The methods were also able to discriminate individuals with PD treated with DBS and levodopa. These methods enable objective characterization and visualization of features extracted from inertial and electromyographic sensors for different groups.
KW - Deep brain stimulation
KW - Discriminant analysis
KW - Electromyography
KW - Inertial sensors
KW - Levodopa
KW - Parkinson's disease
UR - http://www.scopus.com/inward/record.url?scp=85007499591&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85007499591&partnerID=8YFLogxK
U2 - 10.1186/s12938-016-0290-y
DO - 10.1186/s12938-016-0290-y
M3 - Article
C2 - 28038673
AN - SCOPUS:85007499591
SN - 1475-925X
VL - 15
JO - BioMedical Engineering Online
JF - BioMedical Engineering Online
IS - 1
M1 - 169
ER -