Gait Disorder Classification in Neurodegenerative Disorders Using vGRF Data

Gait abnormalities frequently affect older adults, particularly those linked to neurodegenerative disorders like Parkinsons Disease, Huntingtons Disease, and Amyotrophic Lateral Sclerosis. These conditions pose considerable challenges for diagnosis, necessitating prompt detection and intervention, especially in individuals aged 50 and above. This research focuses on the classification of gait disorders using vertical ground reaction force data. Feature extraction was performed using the Continuous Wavelet Transform, and data augmentation techniques were applied to improve dataset diversity and optimize model performance. A deep residual network, ResNet-50, was employed for the classification task, achieving a validation accuracy of 95.05%. The results demonstrate that the proposed model effectively differentiates between neurodegenerative disease cases and healthy individuals. These findings highlight the critical role of integrating vGRF data combined with advanced deep learning techniques to improve the precision of gait disorder classification among older adults,