Enhancing Gait Analysis Through Wearable Insoles and Deep Learning Techniques

This study explores the utilization of wearable sensor insoles for gait signal analysis across diverse participant groups, including elderly individuals, adults, and PD patients. The investigation seeks to establish a connection between changes in stride variability, neurological functionality, and the influence of aging on the human body. To validate the study’s findings, participants’ gait cycles undergo comparison using the Detrended Fluctuation Analysis (DFA) method, which assesses fluctuations in stride time. The results underscore that stride time fluctuations are more pronounced in elderly subjects and PD patients, implying potential irregularities in the fractal characteristics of lower limb dynamics linked to central nervous system control. Another aspect introduced in this chapter outlines an innovative approach to gait assessment, employing the continuous wavelet transform method. This approach addresses the analysis of idiopathic PD severity levels as well as gait variability in age-matched individuals.