Citation: Shin, H. Deep
Convolutional Neural
Network-Based Hemiplegic Gait
Detection Using an Inertial Sensor
Located Freely in a Pocket. Sensors
2022, 22, 1920. https://doi.org/
10.3390/s22051920
Academic Editors: M. Jamal Deen,
Subhas Mukhopadhyay, Yangquan
Chen, Simone Morais, Nunzio
Cennamo and Junseop Lee
Received: 5 December 2021
Accepted: 9 February 2022
Published: 1 March 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the author.
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/).
Article
Deep Convolutional Neural Network-Based Hemiplegic Gait
Detection Using an Inertial Sensor Located Freely in a Pocket
Hangsik Shin
Department of Convergence Medicine, Asan Medical Center, University of Ulsan College of Medicine, 88,
Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Korea; hangsik.shin@gmail.com
Abstract:
In most previous studies, the acceleration sensor is attached to a fixed position for gait
analysis. However, if it is aimed at daily use, wearing it in a fixed position may cause discomfort.
In addition, since an acceleration sensor can be built into the smartphones that people always carry,
it is more efficient to use such a sensor rather than wear a separate acceleration sensor. We aimed
to distinguish between hemiplegic and normal walking by using the inertial signal measured by
means of an acceleration sensor and a gyroscope. We used a machine learning model based on a
convolutional neural network to classify hemiplegic gaits and used the acceleration and angular
velocity signals obtained from a system freely located in the pocket as inputs without any pre-
processing. The classification model structure and hyperparameters were optimized using Bayesian
optimization method. We evaluated the performance of the developed model through a clinical trial,
which included a walking test of 42 subjects (57.8
±
13.8 years old, 165.1
±
9.3 cm tall, weighing
66.3 ± 12.3 kg
) including 21 hemiplegic patients. The optimized convolutional neural network model
has a convolutional layer, with number of fully connected nodes of 1033, batch size of 77, learning
rate of 0.001, and dropout rate of 0.48. The developed model showed an accuracy of 0.78, a precision
of 0.80, a recall of 0.80, an area under the receiver operating characteristic curve of 0.80, and an area
under the precision–recall curve of 0.84. We confirmed the possibility of distinguishing a hemiplegic
gait by applying the convolutional neural network to the signal measured by a six-axis inertial sensor
freely located in the pocket without additional pre-processing or feature extraction.
Keywords:
accelerometry; convolutional neural network; gait analysis; gyroscope; hemiplegia;
inertial signal
1. Introduction
Gait is an important index for evaluating musculoskeletal diseases and degenerative
diseases that increase with aging. The number of studies on gait analysis for disease
diagnosis and daily health management is continuously increasing. Among the various
methods for gait analysis, gait analysis using an inertial sensor, such as a gyroscope or an
accelerometer, has an advantage, since it can be implemented with only a commercially
available small chip. Therefore, the system is simple and easier to use than is a gait analysis
mat or smart insole. In addition, the most recent personal smart and wearable devices are
equipped with an inertial sensor, so it is easy to perform gait analysis without an additional
system. Inertial sensors are widely applied for gait analysis for purposes that range from
detecting gait-specific features [
1
,
2
] and general gait events [
3
–
6
] to clinical applications
in the rehabilitation training monitoring of hemiplegic patients [
7
–
11
] and patients with
Parkinson’s disease [12,13] or Alzheimer’s disease [14].
In our previous study, we developed a wearable acceleration measurement system
that was fixed to the waist to obtain gait data from normal and hemiplegic patients and
classified normal gait and hemiplegic gait using a random-forest classifier based on 165 ex-
tracted features [
7
,
10
]. In most previous studies, the acceleration sensor was attached to
a fixed position to acquire a signal [
6
,
15
,
16
], and gait features were detected for pattern
Sensors 2022, 22, 1920. https://doi.org/10.3390/s22051920 https://www.mdpi.com/journal/sensors
