Research Article

ISSN: 2574 -1241 DOI: 10.26717/BJSTR.2019.17.002951

A Kinect‐Based Portable Automatic Gait Analysis

System: An Experimental Validation

Luigi Pavone1, Gabriele Pasqua2, Paolo Ricciuti1 and Daniele Cafolla1*

1
IRCCS Neuromed, Italy
2
University of Molise, Department of Medicine and Health Science, Italy
*Corresponding author: Daniele Cafolla, IRCCS Neuromed, 86077 Pozzilli, Italy

ARTICLE INFO abstract

Received: April 03, 2019 Gait analysis has established in recent years as an essential tool to quantify
Published: April 17, 2019 rehabilitation effectiveness. Nevertheless, available commercial systems need a structured
environment, are very expensive, require an exact calibration and are not portable.
Citation: Pavone L, Pasqua G, Ricciuti Recently, it has been verified that Kinect can be used for kinematics assessment and has
P, and Daniele C. A Kinect‐Based Port‐ been placed side by side to commercial systems, taking advantage of the potentiality of
able Automatic Gait Analysis System: both systems to provide more detailed analysis. The proposed method aims at having a
An Experimental Validation. Biomed low‐ cost, easy to use, markerless and portable Kinect‐based system that can perform gait
J Sci & Tech Res 17(1)-2019. BJSTR. analysis without needing an accurate calibration and a structured environment.
MS.ID.002951. Keywords: Neuroscience; Neurorehabilitation; Motion Tracking; Kinect-Based System

Introduction
body. Combining both hardware and software gives the Kinect the
Gait analysis has established in recent years as an essential
ability to generate 3D images and recognize human beings within
tool to quantify rehabilitation effectiveness. Gait assessment has
its field of vision [3]. Recent studies [4‐11] showed that Kinect can
become widespread in clinical practice to evaluate and implement
be used for kinematics assessment going through the feasibility
physiotherapy treatments, in order to distinguish between disease
study for clinical use trying to automatize the step detection up to
entities and to customise rehabilitation programs [1].
use commercial system together with the Kinect device to use the
Neurological and neurodegenerative diseases may alter normal advantages of both systems. In this paper, the feasibility of a Kinect‐
gait patterns; for this reason, assessing kinematic and dynamic based, marker less and easy‐to‐use system for gait assessment, by
parameters during deambulation is crucial, in order to detect evaluating the accuracy of various gait parameters extracted by the
impairments underlying reduced motor functions [1]. BTS Gaitlab system as compared to traditional gait analysis systems is evaluated.
(BTS Bioengineering Corp. Quincy, MA 02169 USA) is a complete The methodology is validated emphasizing its advantages and its
system for clinical gait analysis, which integrates kinematic, dynamic low‐cost and easy to use nature.
and electromyographic (EMG) parameters. With its 12 Vicon
Experimental Validation
Infrared Cameras together with two force plats, 2 environmental
cameras and wireless EMG probes, it can be considered as the gold A Microsoft Kinect v2 (sampling frequency = 30 Hz), and 12
standard for gait assessing. BTS system uses different protocols to BTS Bioengineering Vicon cameras (sampling frequency = 120
acquire gait parameters and requires up to 22 markers, fixed on Hz) were used to collect and analyse data. The Kinect camera was
the subject’s body by an expert operator, according to well‐known placed in front of the subject’s path, at 1.2 m height, as illustrated
positioning systems [2]. in Figure 1. During the experimental validation, one healthy subject
(age 37, male, with no neurological and motor deficits) was asked
The Kinect v2 contains three main devices that work together to
to walk straight, covering an area of about 5 meters, delimited
detect subject’s motion and to create the physical representation of
by predefined starting and ending points. The subject performed
the subject on the screen: an RGB colour VGA video camera, a depth
the task 12 times, both for Kinect and Vicon system, for a total of
sensor, and a multi‐array microphone. These components come
24 acquisitions. Since both systems use infrared technology, data
together to detect and track 25 different points on each human

Copyright@ Daniele Cafolla | Biomed J Sci & Tech Res| BJSTR. MS.ID.002951 12552
Volume 17- Issue 1 DOI: 10.26717/BJSTR.2019.17.002951

from Vicon cameras and from Kinect v2 were acquired in two developed using programming languages C# and WPF, using a
different phases to avoid artifacts. A custom application has been third‐party library called Vitruvius [12], which provides useful
developed in order to interface with the Kinect Sensor to detect functionality for analysis of body pose, joints, as well as the rotation
the user’s skeleton and performs all the necessary computations, of a segment in the 3D space and its distance from the Kinect. The
such as determining the joint positions, joint angles; converting application is extremely user‐friendly due to its intuitive graphical
3D coordinates into 2D space (Figure 1). The application was interface, as illustrated in the following Figure 2.

Figure 1: Experimental layout including Kinect v2 mounted on a tripod, and BTS Bioengineering system with twelve infrared
cameras (a). Markers placed on the subject’s body (b).

Figure 2: Kinect interface: measures of joints angles (hip, knee and ankle) along the three axes X,Y,Z.

Before starting the acquisition, it is possible to set up the Name parameters by clicking on “Run MATLAB Analysis” button. Beside
of the patient (that will be used later to create a folder containing these functionalities, the software allows also to see when the
all the outputs provided by the analysis) and the starting leg for the camera actually detects a skeleton and its position in the camera
upcoming walk. Once the acquisition is started (pressing the button view. Once the acquisition starts, every time a new frame is detected
“Start”), it is possible to stop it in any moment (“Stop” button). It is from the Kinect camera a new point will be added to every graph
also possible to clear the previous acquisition and save the current simultaneously and displayed on the interface. Once the subject
section. Once the acquisition ended, it is possible to compute all the has performed the desired task, by pressing the “Stop” button, it

Copyright@ Daniele Cafolla | Biomed J Sci & Tech Res| BJSTR. MS.ID.002951 12553
Volume 17- Issue 1 DOI: 10.26717/BJSTR.2019.17.002951

is possible to save the section in a specific folder containing every computing different parameters as the average value among all
graph data (saved both in jpg and csv format), and a PDF report the sections returning the needed parameters to the software and
containing all the graphs and a video of the performed task. After saving the gait analysis plot and trend to an automatically created
the acquisition the user can, by pressing the “Run MATLAB Analysis” folder for each subject and for each section performed by the latter
button, pass the acquired data to a previously built MATLAB library (Figure 2).
that automatically perform the step detection and the gait analysis,

Results

Figure 3: Knee flexion-extension measured with BTS system and Kinect.

In order to demonstrate that the proposed system provides by the two systems are comparable, as demonstrated by the high
reliable measurements, results provided by the presented system values obtained for R2 and, on the contrary, for the low values
and by BTS system have been compared. Different spatio‐temporal obtained for the RMSE (Figure 3).
parameters and joint angles have been computed and compared,
(Table 1). Conclusion
Table 1: Kinematic measures computed with the two systems: A Kinect‐based, markerless and easy‐to‐use system for gait
the results are presented as mean value across 12 acquisitions assessment has been presented and the accuracy of various gait
parameters extracted by the proposed method as compared to
Parameter BTS system KINECT v2 commercial gait analysis systems has been evaluated. BTS system
has certainly a lot of advantages: it can produce quantitative
Gait cycle 1.08 1.193 information about walking and posture, load anomalies and muscle
duration(s) failure, which would not be measurable with normal clinical exams;
it can with the help of an operator identify the gait cycle events and
Step duration (s) 0.63 0.59
produce a complete clinical report. On the other side, it is extremely
operator‐dependent, both for the positioning of the markers and
Cadence (step/min) 112.52 100.58 also for the post‐processing of the acquired data to create the final
report. Furthermore, data analysis is time‐consuming and requires
Gait cycle length 1.4 1.39
(m) equipment which is not always available in clinical settings.

Mean velocity Similarly, Kinect has also advantaged and drawbacks. Its main
(m/s) 1.3 1.6
limitation is the one camera against the 12 Vicon System Cameras.

For each parameter the mean value across all the acquisitions On the other hand, it is an easy-to-use system; it does not require
has been reported. Figure 3 shows a representative plot of the markers or any kind of preparation of the subject; it does not need
signals computed for a specific joint angle (knee flexion extension) cameras calibration. The experimental campaign demonstrates that
both for BTS system and for the proposed system. For each joint the proposed system provides reliable measurements and present
angle, the Root Mean Square Error (RMSE) and the correlation many advantages compared with the commercially available
coefficient (R2) between the signals extracted with the two systems system “used in this study”. First of all, it is low‐cost, portable and
has been computed. Results showed that measurements provided easy‐to‐use; it does not require markers or any kind of preparation

Copyright@ Daniele Cafolla | Biomed J Sci & Tech Res| BJSTR. MS.ID.002951 12554
Volume 17- Issue 1 DOI: 10.26717/BJSTR.2019.17.002951

of the subject; it does not need cameras calibration, and finally it 6. Wang, Qiang, and Zhanhong Gao (2008) Study on a real‐time image
can automatically compute gait parameters in few seconds giving object tracking system. In 2008 International Symposium on Computer
Science and Computational Technology, pp. 788‐791.
results that are comparable with the already used commercial
systems. A limitation of this study is that it was conducted with 7. Preis J, Kessel M, Werner M, Linnhoff‐Popien C. (2012) Gait recognition
with kinect. In1st international workshop on kinect in pervasive
just one subject, so further studies will involve many subjects in computing New Castle, UK, p. 1-4.
order to demonstrate the reproducibility of the results. Moreover,
8. Auvinet E, Multon F, Aubin CE, Meunier J, Raison M (2015) Detection
other kinematics and spatio‐temporal parameters, which are not of gait cycles in treadmill walking using a Kinect. Gait & posture 41(2):
computed by the proposed system, will be investigated. 722‐725.

References 9. Eltoukhy M, Oh J, Kuenze C, Signorile J (2017) Improved kinect‐based

spatiotemporal and kinematic treadmill gait assessment. Gait & posture
1. Barre A, Armand S (2014) Biomechanical ToolKit: Open‐source 51: 77‐83.
framework to visualize and process biomechanical data. Computer
Methods and Programs in Biomedicine 114(1): 80‐87. 10. Otte K, Kayser B, Mansow Model S, Verrel J, Paul F, et al. (2016) Accuracy
and reliability of the kinect version 2 for clinical measurement of motor
2. (2019) BTS Bioengineering Gaitlab. function. PloS one 11(11): e0166532.
3. (2019) Microsoft. “Kinect v2.” 11. Rocha AP, Choupina HMP, do Carmo Vilas Boas M, Fernandes JM, Cunha
JPS (2018) System for automatic gait analysis based on a single RGB‐D
4. Cafolla D (2019) 3D visual tracking method for rehabilitation path camera. PloS one 13(8): e0201728.
planning. Mechanisms and Machine Science 65: 264‐272.
12. (2019) Vitruvius Kinect framework.
5. Shmuel Springer, Galit Yogev Seligman (2016) Validity of the Kinect for
Gait Assessment: A Focused Review. Sensors (Basel) 16(2): 194.

ISSN: 2574-1241
Assets of Publishing with us
DOI: 10.26717/BJSTR.2019.17.002951
Daniele Cafolla. Biomed J Sci & Tech Res • Global archiving of articles
• Immediate, unrestricted online access
This work is licensed under Creative
• Rigorous Peer Review Process
Commons Attribution 4.0 License
Submission Link: https://biomedres.us/submit-manuscript.php
• Authors Retain Copyrights
• Unique DOI for all articles

https://biomedres.us/

Copyright@ Daniele Cafolla | Biomed J Sci & Tech Res| BJSTR. MS.ID.002951 12555