Comparing inertial measurement units and marker‐based biomechanical models during dynamic rotation of the torso

Abstract

Bronson (2020) Comparing inertial measurement units and markerbased biomechanical models during dynamic rotation of the torso. European

3D trunk orientation measured using inertial measurement units during anatomical and dynamic sports motions

“…Similar RMSE values have been reported in the literature for box lifting and the timed-upand-go walking assessment using a system of heavy, wired IMUs and a logging device to measure the three-dimensional trunk orientation relative to the pelvis. 16,20 Previous studies presented RMSE values lower than 3 degrees 15,18,19,21 or higher than 5 degrees 17 indicating a better or impaired accuracy when compared to results in the present study. These comparisons are limited, however, since several studies considered the segmental orientation (ie, orientation relative to a global coordinate system) of the thorax.…”

“…Recently, RMSE values on average lower than 3 degrees were presented for the trunk orientation relative to the pelvis measured using wireless IMUs. 21 However, in contrast to the present study, the accuracy and similarity measures of each anatomical trunk motion were only provided for the plane of motion.…”

“…The presented R values coincide with previous findings during box lifting, the timed-up-and-go walking assessment, and isolated anatomical trunk motions. 15,16,20,21 Decreased mean R and R 2 values were found out of the plane of motion (lateral flexion and axial rotation angle during flexion/extension and axial rotation angle during lateral flexion). This could be related to a decreased signal-to-noise ratio in the IMUand MOCAP-based trunk angles out of the plane range of motion, 35 especially during flexion/extension (Figure 3; see Appendix 4).…”

“…[15][16][17][18][19][20] However, IMU systems comprised of body-worn wired IMUs, batteries, and logging devices may hamper and limit movement, especially during dynamic sports motions. A few studies evaluated the accuracy of trunk orientation relative to the pelvis measured using wireless IMUs; however, they only examined isolated anatomical trunk motions in the plane of motion, 21 found inaccurate trunk orientation estimates relative to the pelvis during sports motions, 22 examined a sports motion with limited trunk range of motion, 23 or presented limited data and assessed only specific phases of the motion with limited duration. 24 Thus, the accuracy of three-dimensional trunk orientation relative to the pelvis measured using lightweight and wireless IMUs during dynamic sports motions and clinical tasks, such as isolated anatomical trunk motion, remains to be established.…”

Validation of Spatiotemporal and Kinematic Measures in Functional Exercises Using a Minimal Modeling Inertial Sensor Methodology

“…Any error in the initial alignment will be apparent in the mathematical transformation of each individual segment sensor coordinate system to the respective segment coordinate system, with the error compounding where adjoining segments are misaligned. Brice et al [54] demonstrated less agreement between OMC biomechanically-modeled joint angles and un-modeled MARG relative angles than OMC un-modeled relative angles and MARG measured relative angles. This leads to the suggestion that some of the differences in joint angle ROM estimations found in the current study may be due to the differences in modeling assumptions used in each of the OMC and MARG methods and the compounding error occurring throughout the alignment and mathematical transformation process.…”