DocumentCode :
2395031
Title :
A low-cost video-based tool for clinical gait analysis
Author :
Soda, Paolo ; Carta, Alfonso ; Formica, Domenico ; Guglielmelli, Eugenio
Author_Institution :
Med. Inf. & Comput. Sci. Lab., Univ. Campus Bio-Medico di Roma, Rome, Italy
fYear :
2009
fDate :
3-6 Sept. 2009
Firstpage :
3979
Lastpage :
3982
Abstract :
In physical and rehabilitation medicine physicians need to perform clinical gait analysis to assess patients walking ability. Despite the relevant research on motion tracking, gait analysis technologies are far to be commonly diffused in clinical practice since they are quite expensive, need high-structured laboratories and trained personnel who are not always available. In order to overcome such limitations, this work proposes a low-cost, video-based portable tool for clinical gait analysis which provides the bi-dimensional kinematic analysis of walking. The system processes a video stream by means of tracking different markers placed in five anatomical landmarks of the subject´s leg, applying Kalman filter in conjunction with a method that copes with occlusions. The system has been validated on a healthy subject, showing that it is able to reconstruct marker position and leg kinematics even if several occlusions occur.
Keywords :
Kalman filters; biomedical communication; gait analysis; image motion analysis; medical image processing; tracking; video streaming; Kalman filter; bi-dimensional kinematic analysis; clinical gait analysis; leg; low-cost video-based portable tool; marker position; motion tracking; occlusions; physical medicine; rehabilitation medicine; video stream; walking; Algorithms; Automatic Data Processing; Biomechanics; Biophysics; Gait; Humans; Leg; Models, Anatomic; Models, Statistical; Movement; Pattern Recognition, Automated; Software; Video Recording; Walking;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Engineering in Medicine and Biology Society, 2009. EMBC 2009. Annual International Conference of the IEEE
Conference_Location :
Minneapolis, MN
ISSN :
1557-170X
Print_ISBN :
978-1-4244-3296-7
Electronic_ISBN :
1557-170X
Type :
conf
DOI :
10.1109/IEMBS.2009.5333623
Filename :
5333623
Link To Document :
بازگشت