Background and objective: We illustrate a low-cost platform easing the estimation of spatio-temporal parameters (GA-STP) ready for large-scale deployment in fall prevention.
Methods: We used a commercial sensorized carpet with a limited cost and a very coarse resolution. An instrument validation test was accomplished: the Wilcoxon test for paired samples and the correlation test with Spearman method were used to compare the values computed by the platform with reference ones. Hierarchical clustering using Ward's method and ROC curves have been used to assess the performance in a pilot study enrolling patients.
Results: Validation shows no significant difference between computed and reference values of gait speed (ρ-value:0.99; p-value:2.2E-16), step number (ρ-value:0.91; p-value:5.8E-16) and stride-length (ρ-value:0.92; p-value:7.5E-9). The clinical pilot study confirms that the platform may also be used to track the improvements occurring during a rehabilitation program.
Conclusions: We believe that the use of a commercial carpet makes the solution proposed ready to be deployed on a large scale for effectively bringing GA into the clinical practice.
Keywords: Fall prevention; Gait analysis; Remote monitoring,; Sensorized carpet; Signal processing.
Copyright © 2019. Published by Elsevier B.V.