Article
Human Activity Recognition of Individuals with Lower Limb
Amputation in Free-Living Conditions: A Pilot Study
Alexander Jamieson
1
, Laura Murray
1
, Lina Stankovic
2
, Vladimir Stankovic
2
and Arjan Buis
1,
*
Citation: Jamieson, A.; Murray, L.;
Stankovic, L.; Stankovic, V.; Buis, A.
Human Activity Recognition of
Individuals with Lower Limb
Amputation in Free-Living
Conditions: A Pilot Study. Sensors
2021, 21, 8377. https://doi.org/
10.3390/s21248377
Academic Editors: Yangquan Chen,
Nunzio Cennamo, M. Jamal Deen,
Subhas Mukhopadhyay, Simone
Morais and Junseop Lee
Received: 27 October 2021
Accepted: 13 December 2021
Published: 15 December 2021
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2021 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
1
Wolfson Centre, Department of Biomedical Engineering, University of Strathclyde, Glasgow G4 0NW, UK;
alexander.jamieson@strath.ac.uk (A.J.); laura.murray.100@strath.ac.uk (L.M.)
2
Department of Electronic and Electrical Engineering, University of Strathclyde, Glasgow G1 1XW, UK;
lina.stankovic@strath.ac.uk (L.S.); vladimir.stankovic@strath.ac.uk (V.S.)
* Correspondence: arjan.buis@strath.ac.uk
Abstract:
This pilot study aimed to investigate the implementation of supervised classifiers and
a neural network for the recognition of activities carried out by Individuals with Lower Limb
Amputation (ILLAs), as well as individuals without gait impairment, in free living conditions.
Eight individuals with no gait impairments and four ILLAs wore a thigh-based accelerometer and
walked on an improvised route in the vicinity of their homes across a variety of terrains. Various
machine learning classifiers were trained and tested for recognition of walking activities. Additional
investigations were made regarding the detail of the activity label versus classifier accuracy and
whether the classifiers were capable of being trained exclusively on non-impaired individuals’ data
and could recognize physical activities carried out by ILLAs. At a basic level of label detail, Support
Vector Machines (SVM) and Long-Short Term Memory (LSTM) networks were able to acquire 77–78%
mean classification accuracy, which fell with increased label detail. Classifiers trained on individuals
without gait impairment could not recognize activities carried out by ILLAs. This investigation
presents the groundwork for a HAR system capable of recognizing a variety of walking activities,
both for individuals with no gait impairments and ILLAs.
Keywords:
human activity recognition; lower limb amputation; physical activity; machine learning;
lower limb prosthetics
1. Introduction
Following an amputation procedure, Individuals with Lower Limb Amputation
(ILLAs) have been shown to be less physically active than individuals without limb
loss [
1
,
2
], and there has been extensive research into the physical and socio-economical bar-
riers that prevent ILLAs from performing physical activity, such barriers include existing
co-morbidities such as diabetes or chronic obstructive pulmonary disease and a lack of
resources or group support [
3
–
5
]. Maintaining sufficient levels of physical activity is vital to
ILLA’s physical and mental well-being: physical activity has evidently improved heart and
lung functionality and reduced the effects of chronic lower back pain [
4
,
6
]. Physical activity
has further given improved perceptions of the individual’s quality of life, self-esteem,
and body image [
4
,
7
]. Therefore, it is imperative that healthcare professionals can monitor
and evaluate the physical activity of their clients. Modern day healthcare of ILLAs is
limited by a lack of efficient implementation of activity monitoring systems, as evidenced
in a recent literature review [8], and is worthy of further research.
By performing automatic recognition and classification of different activities, it is pos-
sible to establish the basis of a reliable and low-cost activity monitoring system, in which
healthcare professionals that specialise in the care and wellbeing of ILLAs (e.g., physio-
therapists) can track how their patients remain active over long periods of time and in
free-living conditions. Free-living Human Activity Recognition (HAR) studies are not
Sensors 2021, 21, 8377. https://doi.org/10.3390/s21248377 https://www.mdpi.com/journal/sensors
