Gait analysis using wearable sensor

KAVIYARASU M ( M. E ) ANBARASI M : AC20UBM002

SANTHIYA C : AC20UBM023
VISHALI G : AC20UBM037

PRESENTATION TITLE 1
Abstract:
Abnormalities and irregularities in walking (gait) are predictors and indicators of both disease and injury. Gait has traditionally
been monitored and analyzed in clinical settings using complex video (camera-based) systems, pressure mats, or a combination
thereof. Wearable gait sensors offer the opportunity to collect data in natural settings and to complement data collected in
clinical settings, thereby offering the potential to improve quality of care and diagnosis for those whose gait varies from healthy
patterns of movement. This paper presents a gait monitoring system designed to be worn on the inner knee or upper thigh. It
consists of low-power Hall-effect sensors positioned on one leg and a compact magnet positioned on the opposite leg. Wireless
data collected from the sensor system were used to analyze stride width, stride width variability, cadence, and cadence variability
for four different individuals engaged in normal gait, two types of abnormal gait, and two types of irregular gait. Using leg gap
variability as a proxy for stride width variability, 81% of abnormal or irregular strides were accurately identified as different
from normal stride. Cadence was surprisingly 100% accurate in identifying strides which strayed from normal, but variability in
cadence provided no useful information. This highly sensitive, non-contact Hall-effect sensing method for gait monitoring offers
the possibility for detecting visually imperceptible gait variability in natural settings. These nuanced changes in gait are valuable
for predicting early stages of disease and also for indicating progress in recovering from injury.
Introduction:
• WALKING is a fundamental human activity that involves the combined efforts of the muscles, brain
and its nerves. Gait refers to cyclical locomotion achieved through walking. This includes the
movements of arms, legs, hip, feet, and limbs.
• Generally, the gait of each person is unique depending on the gait parameters such as gait phases,
step length and muscle force, etc. Therefore, it helps to understand the individuality and liberty in
humans. The analysis and characterization of gait parameters is called gait analysis.
• Gait analysis helps in investigating different musculoskeletal functions and gait parameters.Gait also
changes as individuals recover from injury or adapt to disability, and monitoring gait during these
processes provides valuable information regarding the success or failure of rehabilitation strategies.
• For example, among children with cerebral palsy, gait is monitored to guide and improve the
viability of adapted walking strategies during rehabilitation training in children.
• Thus, monitoring the simple act of walking can provide important information for detecting, treating,
and healing from injury, for early detection and diagnosis of disease, and for identifying risk of injury
among the elderly and other vulnerable populations.
Objective:

• To detect any abnormalities in locomotion and walking patterns in which an individual's gait is a
combination of complex functions involving use of the body's visual, somatosensory and
vestibular systems.
• To capture and measure information related to the various gait parameters using Hall Effect
Sensors.
• To compare the currently used techniques with the model to be developed on basis of accuracy and
cost.
Literature Survey:
TITLE AUTHOR JOURNAL YEAR METHODOLOGY
SI.NO
1 Gait analysis in Olivier Beauchet, Neuropsychiat 2020 Gait disorders are more prevalent in
demented subjects: Gilles Allali, Gilles ric disease and dementia than in normal aging and are
Interests and Berrut, Caroline treatment conf related to the severity of cognitive decline.
perspectives Hommet, Véronique erence Dementia-related gait changes (DRGC)
Dubost, Frédéric Ass mainly include decrease in walking speed
al provoked by a decrease in stride length and
an increase in support phase.
SI.NO TITLE AUTHOR JOURNAL YEAR METHODOLOGY

2 A review on Delaram IEEE reviews in 2021 Gait analysis has applications in sport,
accelerometry Jarchi, James biomedical computer games, physical rehabilitation,
-based gait Pope, Tracey engineering clinical assessment, surveillance, human
analysis and KM Lee, 11, 177-194 recognition, modeling, and many other fields.
emerging Larisa There are established methods using various
clinical applic Tamjidi, sensors for gait analysis, of which
ations Amirhosein accelerometers are one of the most often
Mirzaei, Saei employed..
d Sanei
SI.NO TITLE AUTHOR JOURNAL YEAR METHODOLOGY

3 Recent Chandra Artificial 2021 Conventionally, the human gait has been
developments in Prakash, Intelligence considered subjectively through visual
human gait Rajesh Review 49 observations but now with advanced
research: Kumar, technology, human gait analysis can be done
parameters, Namita Mitta objectively and empirically for the better
approaches, l quality of life. In this paper, the literature of
applications, the past survey on gait analysis has been
machine learning discussed. This is followed by discussion on
techniques, gait analysis methods
datasets and chall
enges
SI.NO TITLE AUTHOR JOURNAL YEAR METHODOLOGY

4 Human gait Grazia IEEE Journal of 2021 This paper reviews the recent literature on
analysis in Cicirelli, Biomedical and technologies and methodologies for
neurodegene Donato Health Informatics quantitative human gait analysis in the
rative Impedovo, context of neurodegenerative diseases.
diseases: A r Vincenzo The use of technological instruments can
eview Dentamaro. be of great support in both clinical
Roberto diagnosis and severity assessment of these
Marani. pathologies.
Giusee
Pirlo. Tiziana
R
SI. YEA
NO TITLE AUTHOR JOURNAL R METHODOLOGY

5 Gait analysis: MD Journal of 2020 This dependence on cadence (steps per minute) or gait speed
Systems, Akhtaruzzaman, Mechanics underscores the role that assessing cadence can play in overall
technologies, Amir Akramin in Medicine gait monitoring. Studies have shown that while gait speed and
and importance Shafie, Md and Biology step length decrease with age and stride width increases,
Raisuddin Khan cadence remains relatively constant, ranging on average
between 115 and 120 steps per minute for normal gait.

6 Gait Variability Aboutorabi, A.; National There has been growing evidence showing gait variability
2020
in People with Arazpour, M.; library of provides unique information about gait characteristics in
Neurological Bahramizadeh medicine neurological disorders. This study systemically reviewed
Disorders and quantitatively synthesized (via meta-analysis) existing
evidence on gait variability in various neurological diseases,
including Alzheimer's disease (AD), amyotrophic lateral
sclerosis (ALS), cerebellar ataxia (CA), Huntington's
disease (HD), multiple sclerosis (MS), and Parkinson's
disease (PD).
SI.
NO TITLE AUTHOR JOURNAL YEAR METHODOLOGY

7 The Effect of Aboutorabi, National 2020 Gait disorders are common in older adults; prevalence increases
Aging on Gait A.; Arazpour, library of with age and is associated with increased fall risk, mobility
Parameters in M.; medicine decline, and loss of independence. For geriatric patients, the risk
Able-Bodied Bahramizadeh of having gait disorders with an increased fall incidence is even
higher. Consequently, objective gait assessment in the clinics has
become increasingly important for the diagnosis of motor
impairments and the assessment of mobility decline and fall risk.

8 Gait Patterns Lemke, Psychiatric

during over M.R.; 2021 Thus, in many scenarios, cadence serves as a context for
Research
Ground Wendorff, studying other gait parameters. Knowing cadence allows
Locomotion T variations in other gait parameters to be assessed more
in Major accurately. This is not to say, however, that cadence in and of
Depression itself has no value. In fact, among certain physical and
Compared psychological conditions, cadence does indeed change and
with Healthy serve as one of multiple indicators that there may be a
Controls problem. For instance, depressed persons often walk more
slowly
SI.
NO TITLE AUTHOR JOURNAL YEAR METHODOLOGY

9 N. A Study on Gait- Wu, X.; IEEE 2021 Gait monitoring technologies in clinical settings are typically not
Based Parkinson’s Chen, X.; CONFERE worn on the body and are dominated by camera-based systems
Disease Detection Duan, Y.; NCE and force-sensitive platforms. Cameras capture human movement
Using a Force Xu, S.; and then utilize video and image processing to extract gait
Sensitive platform parameters.

10 Gait unsteadiness Jeffrey M BMC In older adults, depression has been associated with
2020
and fall risk in two Hausdorff Psychiatry increased fall risk, but the reasons for this link are not fully
affective disorders: clear. Given parallels between major depression and
a preliminary study Parkinson's disease, we hypothesized that major depression
and related affective disorders would be associated with
impairment in the ability to regulate the stride-to-stride
fluctuations in gait cycle timing.
Existing system:
• The existing system of gait analysis using wearable sensors offers a transformative disadvantage in the real of
healthcare and biomechanics. By leveraging the power of wearable technology, researchers, clinicians, and
therapists gain access to an objective and data-driven approach to understanding human movement.
• Unlike traditional methods that rely on subjective visual assessments or confined laboratory environments,
wearable sensors provide the means to analyze gait in error.
• This allows for a accurate low depiction of how individuals naturally move in their daily lives, offering insights
that are not achievable within the constraints of a controlled setting.

Disadvantage:
• An important disadvantage of gait analysis using wearable sensors is the potential for data accuracy and reliability
issues.
• Disadvantage of gait analysis using wearable sensors is the potential for privacy and data security concerns.
Proposed system:
• This project proposed system both hardware and software design of a gait analysis system. The system proposed is
able to perform gait analysis base on the algorithm developed.
• However, since the magnet sensors can only measure single points, it is unlikely to accurately carry out detail
kinetic analysis, such as moment and power consumption of the joints, which need accurate resultant plantar force
to accomplish.
• This project proposes a wearable gait monitoring system. using hall effect sensors data will send micro controller, a
novel algorithm was developed to collect data from leg movement, using random forest algorithm to detect the gait
problem or not.

Advantage:
• A significant advantage of gait analysis using wearable sensors is the ability to conduct assessments in real-world
environments, providing insights into natural gait patterns during daily activities.
• Another important advantage of gait analysis using wearable sensors is the potential for continuous and long-term
monitoring
• This is especially valuable for rehabilitation programs and tracking disease progression, as it provides a dynamic
view of how gait evolves over time.
Hardware Block diagram:
Portable battery power
supply lithium 3.7 v Software side
One leg magnet

Power booster battery 5v

Collect data leg movement

Hall effect sensor

Time series data

Max – 232 serial reader

data Nano -368 Relay to Random forest
microcontroller Algorithm
LCD Display

Laptop record data and

Detect gait problem
analysis
CIRCUIT DIAGRAM:
CIRCUIT DESCRIPTION:

•The circuit diagram components are power supply lithium , booster, LCD Display, Arduino UNO
microcontroller, half effect sensor and Max232.

•The lithium battery is 3.3v it connected to dc-dc booster it converting 3.3v to 5v. then 5v connected to
controller ground connected to GND.

•Display device is connected to controller digital pins like d2,d3,d4,d,5,d6,d7 respectively. VSS is
connected to GND, and VDD is connected to 5v.

•Hall effect sensor has 3 pin. the positive pin is connected to 5v supply. Out pin connected to digital pin
and GND is connected to ground.
System specifications:

HARDWARE SPECIFICATION
• Nano microcontroller
• Lithium battery
• Regulator
• LCD Display
• Max232
• Hall effect sensor

SOFTWARE SPECIFICATION
• PyCharm
• Arduino idle
• Embedded c
Random forest Algorithm :
•Random Forest is a popular machine learning algorithm that
belongs to the supervised learning technique. It can be used
for both Classification and Regression problems in ML. It is
based on the concept of ensemble learning, which is a
process of combining multiple classifiers to solve a complex
problem and to improve the performance of the model.

•As the name suggests, "Random Forest is a classifier that

contains a number of decision trees on various subsets of the
given dataset and takes the average to improve the predictive
accuracy of that dataset." Instead of relying on one decision
tree, the random forest takes the prediction from each tree
and based on the majority votes of predictions, and it predicts
the final output.
Result:

The work presented here has focused on studying the feasibility of using Hall-effect sensors to
identify gait disturbances. Statistical analyses suggest that leg gap, variability in leg gap, and
cadence offer significant information for differentiating abnormal and ir-regular gait from normal
gait. Future work should capitalize on the results of this work to explore efficient algorithms for
recognizing gait disturbances based on cadence and leg gap data provided by these
wearables.Time series analyses of the entire stride (rather than only the mid-stance peak) should
also be considered in future efforts to explore the full functionality of the Hall-effect wearable in
supporting everyday gait monitoring.
Kit overall setup Serial data
Data logger:
Terminal log file Counter: 12
Date: 10/6/2023 - 10:58:55 AM Counter: 13
----------------------------------------------- Counter: 14
Counter: 1 Counter: 15
Counter: 2 Counter: 16
Counter: 3 Counter: 17
Counter: 4 Counter: 18
Counter: 5 Counter: 19
Counter: 6 Counter: 20
Counter: 7 Time for Count 2 : 7.2s
Counter: 8 Counter: 21
Counter: 9 Counter: 22
Counter: 10
Time for count 1 : 3.4 s -----------------------------------------------
Counter: 11 Date: 10/6/2023 - 10:59:10 AM
End log file
 Sample name Age Of Group No of Steps Time(sec)

Sample 1 4 50 60
(Vedanth)
Sample 2 (Priya) 18 110 60
Sample 3 (Pavithra)
Sample 4 19 101 60
(Mithun) 19 131 60
Sample 5 (Megha)
19 111 60
Sample 6 (Siva) 20 130 60
Sample 7 (Vinay) 21 126 60
Sample 8 (Nivethini)
Sample 9 (Nandhini) 24 121 60
Sample 10 (Hindhu
shree) 25 124 60

23 119 60
Sample 11 (Shakthi ) 30 129 60
Sample 12 (Sai
kumar) 33 131 60
Sample 13
(Muthukumaran) 42 110 60
Sample 14
(Krishnan- Differently 58 70 60
abled)
Sample 15 (Santha) 60 82 60
COMPARATIVE GAIT ANALYSIS:

140

120

Cadence (steps/min)
100

80

60

40

20

0
0 10 20 30 40 50 60 70

Age (in years)

Inference From the graph:

It is inferred from the graph that, though there is an expected decline of cadence as age increases, an evitable
fluctuation is observed when the cadence of late teenage group and mid to late mid ages are compared. It can be
deduced that the changes in lifestyle of individuals in a decade based manner has a huge impact in the health of a
person. These changes include food habits, physical activity and even the intensity and type of medications one intake
throughout their life.
Intake of medications for any ailments in the body indirectly weakens the muscles and bones of the limbs which has a
huge impact in gait pattern irregularities. When the critical point is reached it may even lead to decrease in sensitivity
of areas in the brain and impulse conduction which has a direct impact on gait pattern.
On the other hand, such fluctuations could be as a result of deprived and unstable mental health of the current
generation teenagers. Hence to maintain a healthy gait cycle it is necessary to involve oneself more into works that
demand physical motion and relaxation exercises could be included in the routine to free up the mind and other
elements of the body.
Performance evaluation:
•The study reports a strong accuracy rate of 80% in identifying abnormal or irregular strides and a perfect
82% accuracy in detecting deviations in cadence. This high level of accuracy indicates the system's
effectiveness in distinguishing various gait patterns.

•The system's sensitivity to subtle changes in gait, particularly in cadence and stride width, is a notable
strength. It can detect visually imperceptible gait variations, which is crucial for early disease detection
and recovery monitoring.

•Evaluate the ease of use of the gait monitoring system, both for the individuals wearing it and for
healthcare professionals who may interpret the data. A user-friendly system is more likely to be adopted in
clinical practice.

•Evaluate the safety of the system, including potential risks associated with wearing the sensors and
magnets. Ensure that it does not pose any harm to the individuals being monitored.
Conclusion:

The feasibility of using linear analog Hall-effect sensors for monitoring spatiotemporal gait parameters
has been evaluated. The leg gap and leg gap variability and cadence data collected from these sensors
can successfully detect disturbances in gait within individuals. For the controlled experiments used in
this feasibility study, the temporal parameter of cadence offered 100% accuracy in detecting abnormal
or irregular strides. Based on these initial findings, we can conclude that the Hall-effect sensor offers a
promising low-cost, low-power alternative to classical accelerometer gait disturbance. These Hall-effect
sensor may also act as a supplement to classical systems for continuous monitoring of gait for the
purpose of triggering more comprehensive monitoring by more complex wearables or by traditional
gait analysis systems used in clinical settings. Future work will investigate the design and fabrication of
more compact, integrated Hall-effect wearables, appropriate algorithms for recognizing and flagging
abnormal and irregular gait, and potential integration with other types of sensors that offer information
about other gait parameters including step length, foot orientation, and gait speed.
Reference :
[1] D. T. H. Lai, R. K. Begg, and M. Palaniswami, "Computational intelligence in gait research: a perspective on current applications and
future challenges," IEEE Trans. Inf. Technol. Biomed., vol. 13, no. 5, pp. 687–702, 2020.
[2] E. Dolatabadi, B. Taati, and A. Mihailidis, "An automated classification of pathological gait using unobtrusive sensing technology," IEEE
Trans. Neural Syst. Rehabil. Eng., vol. 25, no. 12, pp. 2336–2346, 2019.
[3] G. Guo, K. Guffey, W. Chen and P. Pergami. "Classification of normal and pathological gait in young children based on foot pressure
data," Neuroinformatics. Vol. 15, no. 1, pp. 13–24, 2019.
[4] P. Sok, T. Xiao, Y. Azeze, A. Jayaraman, and M. V. Albert, "Activity recognition for incomplete spinal cord injury subjects using hidden
Markov models," IEEE Sens. J., vol. 18, no. 15, pp. 6369–6374, 2018.
[5] H.-Y. Lau, K.-Y. Tong, and H. Zhu, "Support vector machine for classification of walking conditions using miniature kinematic sensors,"
Med. Biol. Eng. Comput., vol. 46, no. 6, pp. 563–573, 2019.
[6] [J. D. Farah, N. Baddour, and E. D. Lemaire, "Design, development, and evaluation of a local sensor-based gait phase recognition system
using a logistic model decision tree for orthosis-control," J. Neuroeng. Rehabil., vol. 16, no. 1, p. 22, 2019.
[7] D. Leightley, J. S. McPhee, and M. H. Yap, "Automated analysis and quantification of human mobility using a depth sensor," IEEE J.
Biomed. Health Inform., vol. 21, no. 4, pp. 939–948, 2021.
[8] H. Pham, M. Kawanishi, and T. Narikiyo, "A LLE-HMM-based framework for recognizing human gait movement from EMG," in 2015
IEEE International Conference on Robotics and Automation (ICRA), 2019.
[9] F. Recher, O. Banos, C. D. M. Nikamp, L. Schaake, C. T. M. Baten, and J. H. Buurkc, "Optimizing activity recognition in stroke survivors
for wearable exoskeletons," in 2018 7th IEEE International Conference on Biomedical Robotics and Biomechatronics (Biorob), 2018.
[10] A. de M. e. Souza and M. R. Stemmer, "Extraction and classification of human body parameters for gait analysis," J. Control Autom.
Electr. Syst., vol. 29, no. 5, pp. 586–604, 2018.