Page 83 - Demo

The correlation between knee kinematics and self-reported outcome814Motion data collectionTesting and data collection was performed at the motion lab of UMCG’s Department of Rehabilitation Medicine. Two 40x60 cm force plates (AMTI; Watertown, MA, USA) were integrated in the floor of a 9m long walkway. An optoelectronic motion capture system (VICON MX, Vicon Motion Systems Ltd., Oxford, UK), with 8 cameras sampling at 100Hz was used. 22 14mm spherical markers were assigned and recorded on the lower extremities according to Hayes and Davis.9 One researcher was responsible for marker placement during the entire study. After calibration, joint centers were calculated using VICON Nexus software v2.8 (VICON MX, Vicon Motion Systems Ltd., Oxford, UK). Tests were performed approximately 12 months after ACL reconstruction. Al participants performed three tasks: 1) level walking at a self-selected pace; 2) a single-leg hop for distance (SLH maximum forward jump, jumping and landing on the same leg); and 3) a side jump (maximum sideways jump, jumping from and landing on the same leg). The trials were performed with sports shoes on and hands in free motion. The participants were given 3 practice jumps from which the median distance was obtained to determine the starting distance in front of the force plates. In this way it was ensured that the entire foot lands on the force plate and the participants could familiarize themselves with the task. To determine the starting distance for the side jump, the leg length (greater trochanter tip to lateral malleolus tip) was used. To increase the amount of satisfactory data, three approved trials per task were recorder for each knee. The trials were approved when tasks were performed correctly, i.e. the entire foot landed on the force plate, the landing was stable for at least 3 seconds, and all markers were left in place. Data processingUsing VICON Nexus software v2.8 and an additional custom MATLAB v9.7 script (The MathWorks Inc., Natick, MA, USA), three-dimensional angular displacements and translations in the knee joint were calculated from the position of the markers. Processing and analysis of the data started at initial contact (defined as vertical ground-reaction force >5% body weight) and continued for 200ms. All data was smoothed using the cross-validated quintic spline. The raw 3D marker position data was filtered using a lowMark Zee.indd 81 03-01-2024 08:56