Introduction: The present review aims to provide an overview of the most common uses of wearable inertial sensors in the field of clinical human movement analysis. Areas covered: Six main areas of application are analysed: gait analysis, stabilometry, instrumented clinical tests, upper body mobility assessment, daily-life activity monitoring and tremor assessment. Each area is analyzed both from a methodological and applicative point of view. The focus on the methodological approaches is meant to provide an idea of the computational complexity behind a variable/parameter/index of interest so that the reader is aware of the reliability of the approach. The focus on the application is meant to provide a practical guide for advising clinicians on how inertial sensors can help them in their clinical practice. Expert commentary: Less expensive and more easy to use than other systems used in human movement analysis, wearable sensors have evolved to the point that they can be considered ready for being part of routine clinical routine.

The aim of this study was to validate a wearable inertial measurement unit (IMU), containing a 3D accelerometer and gyroscope, for the estimation of countermovement jump height. The absolute vertical acceleration of the IMU positioned on the back of the participant at L5 level, compensated for trunk rotations, was used to obtain jump height by applying the equation of free-fall to the motion of the IMU. The methodology was tested on 28 participants performing five countermovement jumps each. A reference value for this quantity was obtained using stereophotogrammetry (35.4 cm, s = 4.9). Jump height scores obtained using the proposed methodology (35.9 cm, s = 5.5) presented no significant difference with respect to stereophotogrammetry (P = 0.61). A low bias of 0.6 cm confirmed the accuracy of the estimate, which also showed a high (r = 0.87) and significant (P < 0.0001) correlation with reference values. Furthermore, without compensating accelerations for trunk rotation, jump height was largely underestimated (P < 0.0001) (bias: -12.7 cm) and poorly associated (r = 0.31) with stereophotogrammetry. The results of this study show that the estimation of jump height using inertial sensors leads to accurate results when the measured accelerations are corrected for trunk rotations.