Agreement analysis between Vive and Vicon tracking systems to monitor lumbar postural changes

Introduction/Background Virtual reality (VR) has been shown to be an effective treatment for specific phobias, post-traumatic stress disorder, and chronic pain. Combined with a 3D motion capture system, VR becomes a valuable rehabilitation tool giving the clinician real-time measurements of joint movements. While 3D motion capture is prohibitively expensive, a tracking system released by Vive in March 2017, provides an inexpensive method to provide 6-Degrees of freedom (DOF) motion data. The aim of this study was to assess the accuracy of position and orientation measures reported by Vive Tracker pucks in comparison to Vicon with movements in healthy adults performing trunk motions in a VR game. Material and method Light-reflective markers were mounted on custom designed 3D printed plates along with Vive Tracker pucks. The plates were attached to the thorax and pelvis with elastic straps. Movement of the marker clusters were recorded by twelve Vicon Bonita 10 cameras (spatial resolution of 0.1 mm) using Vicon Tracker and The Motion Monitor software packages. Position and orientation of the Vive Tracker pucks were tracked by two infrared laser emitter units and recorded using the Unity game engine. Both Vicon and Vive data were recorded at 100 Hz. The two data streams were aligned based on game event markers with custom designed Matlab programs. The root meat square error was then calculated across a series of trials. Results The root mean square (RMS) difference between Vicon cluster markers and the Vive Tracker pucks was, on average, 0.26 degrees for the angular displacement and 0.336 mm for the positional data. Conclusion The Vive Tracker pucks provide an accurate measurement of angular displacement and position that is consistent with the gold standard of the Vicon opto-electric system. This finding will help in translating laboratory based VR treatments to the clinic at very little cost.