Machine Learning Based Approach to Identify Neuro-Degenarative Disease using Gait Analysis

Document Type : Primary Research paper


Bannari Amman Institute of Technology, Sathyamanglam


Personal mobility is primarily affected by neurodegenerative diseases.
Characteristics of the disease are stiffness of the muscle and abnormal gait movement.
Impairment of motor activity is a common characteristic indicative of patients with
neurodegenerative (ND) disease, which can disrupt the pathway from the cerebrum to the
muscle and thereby cause movement disorders.In order to differentiate normal gait from
normal gait, we proposed a machine learning based approach. Where we analysed various
classification algorithms and achieved an overall accuracy of 86.35% with 10 features.
Conventional methods including high tech lab setup cameras sensors can be avoided since
the proposed system is effective and wearable. Using manually labeled features, for
algorithms such as K-nearest neighbour, Support Vector Machines, and Decision Trees,
multiple feature sets are used to classify, and the performance of these algorithms is then
recorded. This research introduces a real-time method for the mentioned disorders with an
accuracy of more than 85%.

Volume 12, Issue 3 - Serial Number 3
ICMMNT-2021 International Virtual Conference on Materials, Manufacturing and Nanotechnology, 30th June, 2021.
June 2021
Pages 1705-1711