Automating the Timed Up and Go Test Using a Depth Camera

Abstract

Fall prevention is a human, economic and social issue. The Timed Up and Go (TUG) test is widely used to identify individuals with a high fall risk. However, this test has been criticized because its “diagnostic” is too dependent on the conditions in which it is performed and on the healthcare professionals running it. We used the Microsoft Kinect ambient sensor to automate this test in order to reduce the subjectivity of outcome measures and to provide additional information about patient performance. Each phase of the TUG test was automatically identified from the depth images of the Kinect. Our algorithms accurately measured and assessed the elements usually measured by healthcare professionals. Specifically, average TUG test durations provided by our system differed by only 0.001 s from those measured by clinicians. In addition, our system automatically extracted several additional parameters that allowed us to accurately discriminate low and high fall risk individuals. These additional parameters notably related to the gait and turn pattern, the sitting position and the duration of each phase. Coupling our algorithms to the Kinect ambient sensor can therefore reliably be used to automate the TUG test and perform a more objective, robust and detailed assessment of fall risk.

Evaluation of Home-Based Rehabilitation Sensing Systems with Respect to Standardised Clinical Tests

“…Due to a variety of reasons, including socioeconomic [5], automated solutions for some of these tests have appeared. A significant motivator for the automation of these tests is the elimination of human error [3,[6][7][8].…”

“…With an acknowledged need for automating the TUG and FTSTS tests [5], several sensor-based systems have been proposed in the literature. These are reviewed based on the type of sensors used.…”

“…More recent approaches improve the accuracy of early approaches [24,25] and utilise more advanced webcam sensors such as Kinect [6,25]. These provide better skeleton tracking and are suitable for home use as well as risk-to-fall categorisation [5]. For the FTSTS test, a CCD webcam approach is presented in [26] while in [27] Kinect is used with feedback provided in virtual reality.…”

Predicting TUG score from gait characteristics with video analysis and machine learning

“…First, the subjects are not necessarily asked to perform TUG test even they did in our experiment. We need not try to identify the phases of TUG test as others did [14][15][16] [17]. This makes the method automatic, unobtrusive and easy to deploy at any setting.…”

“…Skeleton data and depth data of Kinect were utilized for detecting six phases of TUG test. Dubois, et al identified the phases of TUG test from the depth data of Kinect camera and additionally, extracted several gait related parameters to classify the subjects as with low or high fall risk [17].…”