| Literature DB >> 26736501 |
Ana Patricia Rocha, Hugo Choupina, Jose Maria Fernandes, Maria Jose Rosas, Rui Vaz, Joao Paulo Silva Cunha.
Abstract
Human motion analysis can provide valuable information for supporting the clinical assessment of movement disorders, such as Parkinson's disease (PD). In this contribution, we study the suitability of a Kinect v2 based system for supporting PD assessment in a clinical environment, in comparison to the original Kinect (v1). In this study, 3-D body joint data were acquired from both normal subjects, and PD patients treated with deep brain stimulation (DBS). Then, several gait parameters were extracted from the gathered data. The obtained results show that 96% of the considered parameters are appropriate for distinguishing between non-PD subjects, PD patients with DBS stimulator switched on, and PD patients with stimulator switched off (p-value <; 0.001, Kruskal-Wallis test). These results are markedly better than the ones obtained using Kinect v1, where only 73% of the parameters are considered appropriate (p-value <; 0.001).Entities:
Mesh：
Year: 2015 PMID： 26736501 DOI： 10.1109/EMBC.2015.7318601
Source DB: PubMed Journal: Conf Proc IEEE Eng Med Biol Soc ISSN： 1557-170X