Comparison of Machine Learning Methods for the Construction of a Standalone Gait Diagnosis Device

In this research, the authors investigate the feasibility of selecting three-dimensional thigh and shank angles as the features of machine learning methods. Four common machine learning techniques, i.e. random forest, k-nearest neighbour, support vector machine and perceptron, were compared in terms...

Full description

Bibliographic Details
Main Authors: Han, Yi Chiew, Wong, Kiing Ing, Murray, Iain
Format: Journal Article
Published: 2020
Online Access:http://hdl.handle.net/20.500.11937/79518
_version_ 1848764067125133312
author Han, Yi Chiew
Wong, Kiing Ing
Murray, Iain
author_facet Han, Yi Chiew
Wong, Kiing Ing
Murray, Iain
author_sort Han, Yi Chiew
building Curtin Institutional Repository
collection Online Access
description In this research, the authors investigate the feasibility of selecting three-dimensional thigh and shank angles as the features of machine learning methods. Four common machine learning techniques, i.e. random forest, k-nearest neighbour, support vector machine and perceptron, were compared in terms of accuracy and memory usage so that a real-time standalone gait diagnosis device can be constructed using low-end inertial measurement units (IMUs). With proper re-sampling and normalisation, they discovered that the support vector machine and perceptron resulted in the top two highest accuracies (96-99%) among the four machine learning methods. The memory requirement of the perceptron is the lowest among the machine learning methods. Therefore, perceptron was selected as the classification algorithm for the standalone gait diagnosis device. The trained perceptron was transferred to the thigh and shank's IMUs to process the data locally in real-time. The constructed standalone gait diagnosis device lit up green or red light emitting diodes when normal or abnormal gaits were detected, respectively. This standalone device was further tested in real-life and achieved a mean classification accuracy of 96.50%.
first_indexed 2025-11-14T11:13:27Z
format Journal Article
id curtin-20.500.11937-79518
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T11:13:27Z
publishDate 2020
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-795182020-09-02T03:53:44Z Comparison of Machine Learning Methods for the Construction of a Standalone Gait Diagnosis Device Han, Yi Chiew Wong, Kiing Ing Murray, Iain In this research, the authors investigate the feasibility of selecting three-dimensional thigh and shank angles as the features of machine learning methods. Four common machine learning techniques, i.e. random forest, k-nearest neighbour, support vector machine and perceptron, were compared in terms of accuracy and memory usage so that a real-time standalone gait diagnosis device can be constructed using low-end inertial measurement units (IMUs). With proper re-sampling and normalisation, they discovered that the support vector machine and perceptron resulted in the top two highest accuracies (96-99%) among the four machine learning methods. The memory requirement of the perceptron is the lowest among the machine learning methods. Therefore, perceptron was selected as the classification algorithm for the standalone gait diagnosis device. The trained perceptron was transferred to the thigh and shank's IMUs to process the data locally in real-time. The constructed standalone gait diagnosis device lit up green or red light emitting diodes when normal or abnormal gaits were detected, respectively. This standalone device was further tested in real-life and achieved a mean classification accuracy of 96.50%. 2020 Journal Article http://hdl.handle.net/20.500.11937/79518 10.1049/iet-spr.2019.0228 restricted
spellingShingle Han, Yi Chiew
Wong, Kiing Ing
Murray, Iain
Comparison of Machine Learning Methods for the Construction of a Standalone Gait Diagnosis Device
title Comparison of Machine Learning Methods for the Construction of a Standalone Gait Diagnosis Device
title_full Comparison of Machine Learning Methods for the Construction of a Standalone Gait Diagnosis Device
title_fullStr Comparison of Machine Learning Methods for the Construction of a Standalone Gait Diagnosis Device
title_full_unstemmed Comparison of Machine Learning Methods for the Construction of a Standalone Gait Diagnosis Device
title_short Comparison of Machine Learning Methods for the Construction of a Standalone Gait Diagnosis Device
title_sort comparison of machine learning methods for the construction of a standalone gait diagnosis device
url http://hdl.handle.net/20.500.11937/79518