Capturing accelerometer outputs in healthy volunteers under normal and simulated-pathological conditions using ML classifiers

V. Filippou, A. C. Redmond, J. Bennion, M. R. Backhouse, D. Wong

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Wearable devices offer a possible solution for acquiring objective measurements of physical activity. Most current algorithms are derived using data from healthy volunteers. It is unclear whether such algorithms are suitable in specific clinical scenarios, such as when an individual has altered gait. We hypothesized that algorithms trained on healthy population will result in less accurate results when tested in individuals with altered gait. We further hypothesized that algorithms trained on simulated-pathological gait would prove better at classifying abnormal activity.We studied healthy volunteers to assess whether activity classification accuracy differed for those with healthy and simulated-pathological conditions. Healthy participants (n=30) were recruited from the University of Leeds to perform nine predefined activities under healthy and simulated-pathological conditions. Activities were captured using a wrist-worn MOX accelerometer (Maastricht Instruments, NL). Data were analyzed based on the Activity-Recognition-Chain process. We trained a Neural-Network, Random-Forests, k-Nearest-Neighbors (k-NN), Support-Vector-Machines (SVM) and Naive Bayes models to classify activity. Algorithms were trained four times; once with 'healthy' data, and once with 'simulated-pathological data' for each of activity-type and activity-task classification. In activity-type instances, the SVM provided the best results; the accuracy was 98.4% when the algorithm was trained and then tested with unseen data from the same group of healthy individuals. Accuracy dropped to 52.8% when tested on simulated-pathological data. When the model was retrained with simulated-pathological data, prediction accuracy for the corresponding test set was 96.7%. Algorithms developed on healthy data are less accurate for pathological conditions. When evaluating pathological conditions, classifier algorithms developed using data from a target sub-population can restore accuracy to above 95%.Clinical Relevance - This method remotely establishes health-related data of objective outcome measures of activities of daily living.

Original languageEnglish
Title of host publication42nd Annual International Conferences of the IEEE Engineering in Medicine and Biology Society
Subtitle of host publicationEnabling Innovative Technologies for Global Healthcare, EMBC 2020
PublisherIEEE
Pages4604-4607
Number of pages4
ISBN (Electronic)9781728119908
DOIs
Publication statusPublished - 27 Aug 2020
Event42nd Annual International Conferences of the IEEE Engineering in Medicine and Biology Society, EMBC 2020 - Montreal, Canada
Duration: 20 Jul 202024 Jul 2020

Publication series

NameProceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
Volume2020-July
ISSN (Print)1557-170X

Conference

Conference42nd Annual International Conferences of the IEEE Engineering in Medicine and Biology Society, EMBC 2020
Country/TerritoryCanada
CityMontreal
Period20/07/2024/07/20

Bibliographical note

©2020 IEEE

Cite this