Metrological Characterization of a Wearable Device for the Assessment of Gait Parameters

Wearable devices are widely marketed for sports applications, particularly inertial measurement unit (IMU) sensors, which playa pivotal role in enhancing the athletes' performance. IMU sensors can capture gait dynamics, allowing real-time monitoring during training and competition. Positioning these sensors on the upper torso holds a distinctive advantage, as it allows researchers and coaches to attain accurate data regarding athletes' motion. This research focuses on the metrological analysis of the K-AI Wearable Tech device by K-Sport (Fano, Italy), with a specific emphasis on the stride duration. The study involves a comparison of the device measurement results with those obtained from single point pressure sensors (reference device, gold standard). The results demonstrate high measurement precision and accuracy. The distribution of measurement differences (i.e., residuals) shows a narrow spread with low mean value (approximately 0 s) and low standard deviation (0.01 s and 0.15 s for left and right feet, respectively). Bland-Altman plots confirm agreement with reference device and the Pearson's correlation coefficient shows a strong correlation (0.98 and 0.97 for left and right foot, respectively). Hence, the wearable device can be considered suitable for gait analysis in sport applications, where it results particularly comfortable for upper torso positioning, not interfering with the athlete's movements.