MARKER REGISTRATION FOR INVERSE KINEMATIC MODELS OF THE UPPER LIMB: IMPORTANT CONSIDERATIONS FOR THE SPORT SCIENTIST
Keywords:cricket, statistical parametric mapping, accuracy, reliability
AbstractThe scaling of rigid-linked skeletal models is an important consideration for researchers looking to calculate joint angles via inverse kinematics (IK). It has been suggested (Dunne et al., 2013) that registering marker positions with known kinematics during scaling can improve the accuracy of IK derived lower limb joint angles during gait. The purpose of this manuscript was to determine if registering marker positions with known joint kinematics can improve the accuracy and reliability of time varying IK derived elbow flexion/extension (FE) estimates during cricket bowling. Registering marker positions and joint kinematics (MKR) resulted in improved accuracy than marker positions only (MR) (RMSE = 8.9° v 25.1°) when compared with known DK derived elbow angles. The inter-tester reliability of MKR model elbow extension range was also superior (ICC = 0.626 v 0.318).