Assessing Pediatric Gait Symmetry Through Accelerometry and Computational Intelligence

Abstract

This paper focuses on the use of wearable sensors to acquire and process motion data, which is essential for monitoring physiological movement and identifying gait disorders. It is particularly relevant in pediatrics, neurology, and rehabilitation. The research evaluates body motion symmetry in children using accelerometric data, taking into account factors such as age, diagnosis, and gender. Signals were recorded from 35 children (average age 10.8 years) using mobile sensors and were analyzed using digital signal processing techniques and classification methods. The proposed methodology includes data acquisition by smartphone sensors, wireless data export to a remote drive, and data processing through a graphical user interface. The highest classification accuracy of walking features, at 92.0%, was achieved with a two-layer neural network. The findings underscore the effectiveness of these tools in rehabilitation, fitness monitoring, and neurological studies.