Quantification of the Human Postural Control Using the Nonlinear Analysis of Cop Variations during the Quiet Standing
Farshad Samaei1, Maryam Daneshfar2, Samane Safari Beydokhti3
1Farshad Samaei, Department of Biomedical, Mashhad Branch, Islamic Azad University, Mashhad, Iran.
2Maryam Daneshfar, Department of Biomedical, Mashhad Branch, Islamic Azad University, Mashhad, Iran.
3Samane Safari Beydokhti, Department of Biomedical, Mashhad Branch, Islamic Azad University, Mashhad, Iran.
Manuscript received on November 25, 2013. | Revised Manuscript received on December 15, 2013. | Manuscript published on December 30, 2013. | PP: 57-60 | Volume-3, Issue-2, December 2013. | Retrieval Number: B2355123213/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The aging is an effective factor on the quality of standing in healthy subjects. Some neural disorders, degrades the quality of standing, so that the quality of standing in young patient may be as well as the quality of standing in elderly healthy subject. So in this study, the subjects were divided to five age groups, and the age group the subject belonging to it is the measure to quantify the quality of postural control. The subjects were aged between 25-75 years old. The Center of Pressure (CoP) position variations and Center of Pressure (CoP) position velocity during the quiet standing were analyzed through the RQA (Recurrence Quantification Analysis) method. The extracted nonlinear features were fed to the nonlinear classifiers, and the output of classifiers specified the age group which each subject belongs to it. The SVM, MLP neural network, and RBF neural network were the used classifiers. In this manner, the quality of subject postural control could be quantified between1 to 5. Results show the SVM classifier with polynomial kernel reached the best performance of 97.44% accuracy.
Keywords: Quiet Standing, Quantification, Aging, RQA, Nonlinear classification.