Abstract
The understanding of human movement has been greatly enhanced through the use of kinematics, and it has provided a more sensitive metric to evaluate recovery following a stroke. Conventional acquisition techniques (e.g., Vicon, Optotrak) typically remain non feasible clinical options as they are immobile, expensive, and require long setup times. One alternative is using accelerometers, however movements recorded via accelerometers may be difficult to interpret as the axes shift with rotational movements. Advancements in sensor technologies have enabled kinematic acquisition without markers and portability, while effectively reducing costs. One such device is the LEAP sensor which enables the specific kinematic capture the hands and digits. Since this is a novel technology there remain questions regarding the validity of the kinematic measures and the feasibility of clinical use. Thus the purpose of this study was to test the validity and accuracy of LEAP compared to a gold standard system (Optotrak). Participants positioned their index finger at 15 different targets (located in vertical and horizontal planes on a monitor). Positional data was acquired using both systems. We compared the target coordinates of the two systems using correlations and measures of variability (constant error (CE), variable error (VE)). Our results demonstrate significant correlation between the systems (0.99 for horizontal and 0.95 for vertical direction) but the LEAP did not have the same precision (CE) as Optotrak; however variability (VE) was comparable. Overall these results suggest that the LEAP system could be adopted as a clinically feasible option to conventional kinematic capture systems.
