Search

A promising but still scarcely explored strategy for the estimation of gait parameters based on inertial sensors involves the adoption of machine learning techniques. However, existing approaches are reliable only for specific conditions, inertial measurements unit (IMU) placement on the body, protocols, or when combined with additional devices. In this paper, we tested an alternative gait-events estimation approach which is fully data-driven and does not rely on a priori models or assumptions. High-frequency (512 Hz) data from a commercial inertial unit were recorded during 500 steps performed by 40 healthy participants. Sensors' readings were synchronized with a reference ground reaction force system to determine initial/terminal contacts. Then, we extracted a set of features from windowed data labeled according to the reference. Two gray-box approaches were evaluated: (1) classifiers (decision trees) returning the presence of a gait event in each time window and (2) a classifier discriminating between stance and swing phases. Both outputs were submitted to a deterministic algorithm correcting spurious clusters of predictions. The stance vs. swing approach estimated the stride time duration with an average error lower than 20 ms and confidence bounds between ±50 ms. These figures are suitable to detect clinically meaningful differences across different populations. [ABSTRACT FROM AUTHOR]

Plantar pressure measurements provide useful information to diagnose a diverse range of foot disorders; unfortunately, the commercially available measurement systems are undesirably sensitive to several disturbances, but this aspect is mostly neglected in the literature. This paper describes the results of an experimental campaign aiming at the identification of pressure measuring system metrological performances, at system modeling, and at the implementation of correction procedures. The sensor model was implemented using the results of static and dynamic tests performed on a pedar-X plantar pressure measurement system. The static calibration was performed by analyzing the effect of temperature, single sensor coverage area, local curvature, tangential forces, long-term stability (creep), and sensor crosstalk on the system performances. The dynamic calibration was performed on an electrodynamic shaker, identifying the single sensor frequency response function and the hysteresis under different average loads. The dynamic sensor model is based on the Kelvin–Voigt model, which is representative of the viscoelastic behavior of the material. The model allowed us to compensate both the creep (i.e., the behavior under static loads) and the nonunitary frequency response function. A deconvolution-based algorithm has been proposed to compensate the sensor crosstalk effects, although its implementation requires additional investigations. Experimental results of bobbing and gait tests showed that, with the adoption of the proposed compensation algorithms, the force and center of pressure errors could be reduced by more than 50\% of their initial values. [ABSTRACT FROM PUBLISHER]