Repeatability of measuring knee flexion angles with wearable inertial sensors

As assessment with inertial-measurement-units (IMUs) increases in research and in clinics, it is important to be aware of the repeatability of these sensors. The objectives of this experiment were to evaluate the measurement repeatability of IMU joint angles using a repeatable robot controller and an anthropomorphic leg phantom and to determine effects of joint speed and sensor positioning on the angles collected by these sensors. Comparisons to an electro-goniometer and three-dimensional (3D) motion capture cameras were also completed. Two dual-IMU setups (posterior and lateral) were tested concurrently with an electro-goniometer and 3D motion capture cameras using a repeatable robot controller and a leg phantom. All modalities were attached to the phantom, which was flexed 10 times using a pre-programmed motion pathway during each test. Mean angles were compared across tests. Effects of joint speed, sensor re-positioning, and anatomical placement of the sensors on repeatability were assessed. Re-positioning caused greater deviation to the maximum and minimum angles than differences in speed. Overall, the means ± standard deviations, and 95% confidence intervals of the maximum angles across all tests for the 3D camera markers, electro-goniometer, posterior IMUs, and lateral IMUs were 119.4 ± 0.3° (119.4, 119.5), 112.4 ± 0.5° (112.3, 112.5), 116.2 ± 2.4° (115.7, 116.7), and 118.3 ± 1.1° (118.1, 118.6). Both posterior and lateral IMU setups demonstrated acceptable repeatability in measurement of range of motion that was advantageous to manual goniometer methods. Posterior and lateral IMU setups demonstrated overlapping standard deviations about their means.