Static and dynamic accuracy of a magnetic-inertial measurement unit used to provide racket swing kinematics

Magnetic-inertial measurement units (MIMUs) are becoming more prevalent in sports biomechanics and may be a viable tool to evaluate kinematic parameters. This study examined the accuracy of a MIMU to estimate orientation angles under static conditions and dynamically from a squash racket during a forehand drive shot. A MIMU was mounted onto a goniometer and moved through 0-90 degrees, with static data collected at 10 degrees increments during 10 repetitions of all three axes. Typical error analyses showed the MIMU to be very reliable (TE <= 0.03 degrees). MIMU accuracy was determined via intraclass correlation coefficients (ICC) (r > 0.999, p < 0.001). An ordinary least products regression showed no proportional bias and minimal fixed bias for all axes. Dynamic accuracy was assessed by comparing MIMU and optical motion capture data of squash racket swing kinematics. A MIMU was fixed onto a racket and 10 participants each hit 10 forehand shots. Mean orientation angle error at ball impact was <0.50 degrees and ICC showed very high correlations (r >= 0.988, p < 0.001) for all orientations. Swing phase root mean squared errors were <= 2.20 degrees. These results indicate that a MIMU could be used to accurately and reliably estimate selected racket swing kinematics.