A 3D camera based automated T-GUGT tes

Bart Jansen, Rudi Deklerck

Research output: Contribution to journalMeeting abstract (Journal)

Abstract

Background and Aims: In the timed Get Up and Go Test (T-GUGT)
to total time required to rise from a chair, walk 3m., turn, walk
back and sit down again, is compared to a threshold. This test is
widely used in fall risk and mobility assessment. The total time is
measured by an observer and hence prone to errors. The durations
of the different phases are typically not available. This small study
aims to investigate whether image processing methods applied to
3D camera images could be applied to automate the T-GUGT test.
Methods: Custom image processing, calibration and pose transition
methods were developed and applied to 3D camera recordings of
4 healthy controls, each performing 6 consecutive instances of the
T-GUGT. The durations of the different phases of the test and the
total time taken were compared using Pearson’s Correlation to
the respective durations obtained from manually annotating the
recorded images at the frame level.
Results: The calculated total duration of the test correlates well
to the ground truth (R2 = 0.761), while the correlation between the
durations of the individual phases ranges from 0.69 for the forward
walking phase down to 0.2 for sitting down and rising from the
chair.
Conclusions: The initial results suggest that a camera based
automatic T-GUGT test is feasible, but that more research followed
by proper validation is required. The added value of the proposed
approach is that it additionally provides durations of the individual
phases.
Original languageEnglish
Pages (from-to)26-26
JournalParkinsonism and Related Disorders
Volume16
Issue numbers1
Publication statusPublished - 26 Feb 2010
Event3rd International Congress on Gait & Mental Function - Washington, United States
Duration: 26 Feb 201028 Feb 2010

Keywords

  • e-health
  • timed up and go test

Fingerprint

Dive into the research topics of 'A 3D camera based automated T-GUGT tes'. Together they form a unique fingerprint.

Cite this