Comparison of 2-D and 3-D Analysis of Running Kinematics and Actual Versus Predicted Running Kinetics

Background Providing clinicians with an accurate method to predict kinetic measurements using 2D kinematic motion analysis is crucial to the management of distance runners. Evidence is needed to compare the accuracy of 2D and 3D kinematic measurements as well as measured and estimated kinetic variables. Purposes The objectives of this study were to (1) compare 2D video analysis of running kinematics with gold standard 3D motion capture and, (2) to evaluate published equations which estimate running kinetics using 2D kinematic and spatiotemporal values and modify these equations based on study findings. Design Controlled laboratory study, cross-sectional design Methods Runners who averaged at least 20 miles per week were invited to participate. Athletes ran on an instrumented treadmill at their preferred training pace for a 6-minute warm-up. Markers were placed over designated anatomical landmarks on both sides of the pelvis as well as the left lower extremity. Subjects then ran at their preferred speed and kinematic data were recorded using both the 2D and 3D camera systems at 240 frames/second. Additionally, ground reaction forces were recorded at 1200Hz. 2D and 3D kinematic values were compared and published kinetic prediction formulas were tested. Linear regression was used to develop new prediction equations for average loading rate (AVG_LR), peak vertical ground reaction force (VERT_GRF), and peak braking force (PK_BRK). Paired t-tests were used to assess differences between the 2D and 3D kinematic variables and the measured (MEAS) and calculated (CALC) kinetic variables. Results Thirty runners (13 men and 17 women) voluntarily consented to participate in this study and the mean age of the participants was 31.8 years (range 20 to 48 years). Although significant differences existed, all 2D kinematic measures were within 2°-5° of 3D kinematic measures. Published prediction equations for AVG_LR and VERT_GRF were supported, but new prediction equations showed higher R2 for AVG_LR (0.52) and VERT_GRF (0.75) compared to previous work. A new prediction equation for PK_BRK was developed. No significant differences were found between the MEAS and CALC kinetic variables using the new equations. Conclusion Accurate predictions of kinetic variables can be made using spatiotemporal and 2D kinematic variables. Level of Evidence Level 2