Title
Automated analysis of gait and modified timed up and go using the Microsoft Kinect in people with Parkinson's disease: associations with physical outcome measures.
Abstract
Instrumenting physical assessments in people with Parkinson's disease can provide valuable and sensitive information. This study aimed to investigate whether variables derived from a Kinect-based system can provide incremental value over standard habitual gait speed (HGS) and timed up and go (TUG) variables by evaluating associations with (1) motor and (2) postural instability and gait difficulty (PIGD) subscales of the Unified Parkinson's Disease Rating Scale (UPDRS). Sixty-two individuals with Parkinson's disease (age 66 ± 7 years; 74% male) undertook an instrumented HGS and modified TUG tests, in addition to the UPDRS. Multivariable regression models were used to evaluate the associations of the Kinect measures with UPDRS motor and PIGD scores. First step length during the TUG and average step length and vertical pelvic displacement during the HGS were significantly associated with the PIGD subscale (P < 0.05). The only Kinect-derived variable showing additive benefits over the standard measures for the PIGD association was HGS vertical pelvic displacement. The only standard or Kinect-derived variable significantly associated with the motor subscale was first step length during the TUG (P < 0.01). This study provides preliminary evidence to support the use of a low-cost, non-invasive method of instrumenting gait and TUG tests in people with Parkinson's disease. Graphical abstract ᅟ.
Year
DOI
Venue
2019
10.1007/s11517-018-1868-2
Medical & biological engineering & computing
Keywords
Field
DocType
Parkinson’s disease,Assessment,Gait,Instrumentation,Kinect
Computer vision,Parkinson's disease,Gait,Postural instability,Regression analysis,Rating scale,Artificial intelligence,Physical medicine and rehabilitation,Gait difficulty,Mathematics
Journal
Volume
Issue
ISSN
57
2
1741-0444
Citations 
PageRank 
References 
1
0.35
0
Authors
11