Please use this identifier to cite or link to this item:
Title: Parkinson's Disease Assessment Based on Gait Analysis Using an Innovative RGB-D Camera System
Authors: Rocha,AP
Hugo Miguel Choupina
João Paulo Cunha
Issue Date: 2014
Abstract: Movement-related diseases, such as Parkinson's disease (PD), progressively affect the motor function, many times leading to severe motor impairment and dramatic loss of the patients' quality of life. Human motion analysis techniques can be very useful to support clinical assessment of this type of diseases. In this contribution, we present a RGB-D camera (Microsoft Kinect) system and its evaluation for PD assessment. Based on skeleton data extracted from the gait of three PD patients treated with deep brain stimulation and three control subjects, several gait parameters were computed and analyzed, with the aim of discriminating between non-PD and PD subjects, as well as between two PD states (stimulator ON and OFF). We verified that among the several quantitative gait parameters, the variance of the center shoulder velocity presented the highest discriminative power to distinguish between non-PD, PD ON and PD OFF states (p = 0.004). Furthermore, we have shown that our low-cost portable system can be easily mounted in any hospital environment for evaluating patients' gait. These results demonstrate the potential of using a RGB-D camera as a PD assessment tool.
metadata.dc.type: conferenceObject
Appears in Collections:C-BER - Articles in International Conferences

Files in This Item:
File Description SizeFormat 
  Restricted Access
1.11 MBAdobe PDFView/Open Request a copy

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.