Are Wearable Sensors Valid and Reliable for Studying the Baseball Pitching Motion? An Independent Comparison With Marker-Based Motion Capture

Abstract

Background: In recent years, the prevalence of medial ulnar collateral ligament injuries has increased in throwers of all ages and skill levels. The motusBASEBALL sensor possesses an inertial measurement unit (IMU) that has been developed and applied to the throwing arm to allow for measurements of several objective parameters, which may prove beneficial for monitoring, rehabilitation, and injury prevention in the throwing athlete. However, the reliability, consistency, and validity of the IMU have not been independently assessed. Purpose: To evaluate the reliability, consistency, and validity of the motusBASEBALL sensor compared with the historic gold standard of marker-based motion capture. Study Design: Controlled laboratory study. Methods: A total of 10 healthy male baseball athletes with varsity-level high school experience volunteered to participate in this study. Participants were fitted with 37 retroreflective markers for motion capture and the motusBASEBALL IMU sensor. Participants threw 5 fastballs at maximum effort, with measurements recorded simultaneously by motion capture and the IMU. Arm slot, arm speed, arm stress, and shoulder rotation were measured and compared. Results: Of the 4 metrics generated by the IMU, significant differences were found for 3 of the throwing metrics compared with motion capture including arm slot (5.0°± 6.1°; P = .037), elbow varus torque (9.4 ± 12.0 N·m; P = .037), and shoulder rotation (6.3°± 6.1°; P = .014). Arm speed did not demonstrate a statistically significant difference (29.2 ± 96.8 rpm; P = .375). The IMU consistently underreported pitching performance values. Shoulder rotation exhibited excellent reliability with <5° of error, and arm slot demonstrated good reliability with <10° of error. Arm stress and arm speed were less reliable. Conclusion: The IMU was not accurate or valid for arm slot, arm stress, and shoulder rotation compared with marker-based motion capture. It was relatively accurate for arm speed. Despite its lack of validity, it was consistent and reliable for arm speed and shoulder rotation and relatively reliable for arm slot and arm stress. Caution should be used when comparing values provided by this IMU to the gold standard of marker-based motion capture. Clinical Relevance: IMU technology has potential to be used in monitoring, rehabilitation, and injury prevention in throwing athletes if valid. This study demonstrates that the values provided by the IMU should not be considered equivalent to those generated by the gold standard of marker-based motion capture; however, there may still be a role for this technology when relying on its internal consistency for intrathrower comparisons and tracking.