Conference PaperPDF Available

Detecting and Minimizing Bad Posture using Postuino Among Engineering Students

Conference Paper

Detecting and Minimizing Bad Posture using Postuino Among Engineering Students

Abstract

To understand how good posture minimizes computer related injury's pain, we developed Postuino. A device that warns the computer users if they are leaning too close towards computer screen. Postuino has an accompanying web application that visualizes the collected data and displays a chart to simplify comparing between straight time and slouch time. Also, the app suggests taking frequent breaks to minimize the risk of injuries and to increase productivity. Then, we designed an experiment with different factors to evaluate the efficiency of Postuino. In our study, 24 subjects first use the computer for 3 hours after disabling Postuino's alert system. Afterwards, they use the computer again for two more 3 hour phases after enabling the alert system. We collected data, analyzed it, and presented the results in this paper.
Detecting and Minimizing Bad Posture using Postuino Among Engineering Students
Reem Alattas and Khaled Elleithy
Department of Computer Science & Engineering
University of Bridgeport
Bridgeport, CT, USA
ralataas@my.bridgeport.edu, elleithy@bridgeport.edu
Abstract— To understand how good posture minimizes
computer related injury’s pain, we developed Postuino. A
device that warns the computer users if they are leaning too
close towards computer screen. Postuino has an accompanying
web application that visualizes the collected data and displays
a chart to simplify comparing between straight time and slouch
time. Also, the app suggests taking frequent breaks to
minimize the risk of injuries and to increase productivity.
Then, we designed an experiment with different factors to
evaluate the efficiency of Postuino. In our study, 24 subjects
first use the computer for 3 hours after disabling Postuino’s
alert system. Afterwards, they use the computer again for two
more 3 hour phases after enabling the alert system. We
collected data, analyzed it, and presented the results in this
paper.
Keywords— embedded systems, microcontrollers, Arduino,
ultrasonic sensors, posture, experimentation.
I. INTRODUCTION
Computers are considered as an essential part of our daily
life. Nowadays, computers are used in different environment;
including home, school, and work. Although computers
made our lives easier and more efficient, there were some
injuries related to computer use; such as muscle and joint
pain. Using a computer for long periods often results in
postural compromises such as kyphosis, i.e. rounding of the
upper back, short hip flexors and quadriceps, and forward
head position. These in turn lead to head, neck and shoulder
tension and pain, lower back pain, improper breathing
patterns, nerve compression, and increased potential for disc
herniation and arthritis [1]. These problems are caused by
bad posture and made worse by sitting for long periods of
time. Therefore, we developed Postuino to detect bad posture
among computer users and suggest short breaks after sitting
and using the computer for a certain amount of time.
A survey of 1,544 graduating seniors at Harvard
University, reported that over 50% of the students
experienced symptoms with computer use, and 12.6%
experienced symptoms after computing for 1 hour or less [2].
Risk factors were academic concentration in computer
science, female gender, and using a computer more than 20
hour/week. Compared to undergraduate students, graduate
students were at greater risk for musculoskeletal symptoms
and disorders due to the intensive computer use required for
data analysis and thesis writing, as well as employment as
graduate student researchers and teaching assistants [3].
Another survey of 304 engineering graduate students at
Harvard University reported that 65% of those students
experienced persistent pain in upper extremity or neck during
graduate school. Of all participants, 60% reported recurrent
pain was related to computer use.
As computer adoption increases we can expect a
corresponding increase in the occurrence of neck pain if
appropriate countermeasures are not employed. Thus, we
developed Postuino to detect bad posture by measuring the
distance between the user and computer monitor in order to
indicate if the user is leaning too close to the computer
monitor. Postuino alerts the user to prevent slouching and
suggests short breaks to minimize the impact of too much
sitting and staring at the computer.
Although sitting requires less muscular effort than
standing, it still causes physical fatigue because it holds parts
of human body steady for long periods of time. This reduces
circulation of blood to muscles, bones, tendons and
ligaments; which leads to stiffness and pain [4].
In this paper, we propose Postuino, a novel solution for
fixing bad posture and reducing muscle pain consequently.
Section 2 discusses previous work done in the field of
posture detection. Section 3 describes the new tool, its
different parts, and the accompanying web application. Then,
an experiment is discussed and analyzed in section 4.
Finally, section 5 evaluates the new tool and concludes the
paper.
II. RELATED WORK
Bad posture detection is a field where extensive
development has been done. Many devices were developed
to detect and correct poor posture such as; iPosture, BASN,
and Lumo Back. However, all these devices belong to
wearable technology arena. Some of those devices will be
discussed next.
iPosture is a small 1-inch round device designed to be
worn by the user, detect bad posture, and notify the wearer
with a vibration. It warns the user when detecting any
deviation greater than three degrees from the chosen posture
that lasts longer than one minute. The chosen posture is
saved by pushing a button [5]. It is convenient to have such a
small posture detection device, though it is uncomfortable to
wear all the time. The chosen posture as well might not be
the best posture to be kept all the time. This device does not
give any feedback to the user about the posture performance
during the time where that tool was used.
Body Area Sensor Network (BASN) is a solution for
wireless and wearable posture recognition based on a
custom-designed wireless body area sensor network, called
WiMoCA. Nodes of the network, mounted on different parts
of the human body, exploit tri-axial accelerometers to detect
body postures [6]. The disadvantages of that solution lies in
the impracticality of wearing many sensors and purring daily
life activities such as working or studying and the lack of
feedback provided for users. On the other hand, one of the
challenges this tool faces is power consumption
optimization.
Lumo Back is a comprehensive bio-mechanical posture
sensor and body movement feedback system that monitors
the wearer’s posture and coaches to improve throughout the
day. It is considered one of the best available solutions,
because it is only 8.5 mm thick. The sensor gently vibrates
when the user slouches, which is similar to iPosture alert
methodology. When synced with a compatible mobile
device, Lumo Back keeps track of the user’s daily activities;
such as steps taken, time spent sitting, calories burned, and
sleep habits [7]. The downside of Lumo Back is the lack of a
built in adjustor for variations in different users’ “correct
postures,” since each person has a different curvature to their
spine, their “correct posture” may be different than someone
else’s.
All the above devices are worn by the user and target
general population without focusing on the reasons that
caused bad posture. Postuino aims to detect bad posture for
computer users while using their computers. It is the only
posture detector device that is not worn. Instead, it can be
positioned next to the computer monitor or mounted on top
of the monitor depending on the users’ preferences. Postuino
provides valuable feedback to the users and suggests short
breaks to avoid sitting for too long as we will see in the next
sections.
III. POSTUINO
Postuino warns computer users if they are leaning too
close towards the computer screen, by measuring the
distance between the user and computer monitor in order to
indicate if the user is leaning too close towards the computer
monitor. Also, it provides excellent feedback for users about
the time they spend sitting and slouching. This information
can be used by the users themselves or their physicians or
chiropractors for health monitoring purposes. In the next
subsections, we will elaborate on the hardware used to make
the prototype, distance measurement algorithm, data
visualization tool used to convert data to graphical
representation, and the accompanying web application to
display the visualized data for each user.
A. Hardware
Our design uses Arduino Micro and ultrasonic sensor. It
can be placed next to the computer facing the user, as shown
in Fig. 1, or mounted on top of the computer monitor, as
shown in Fig. 2. When the user leans towards the computer
monitor, the distance between him/her, the computer, and the
Postuino accordingly falls below a certain threshold. Then,
an LED lights up and Piezo speaker plays chosen melody in
order to alert the user to correct his/her posture.
We chose to use Arduino Micro because of its small size
compared to other Arduino boards. Moreover, Arduino
Micro has pins that can be connected straight onto a
prototyping breadboard, which allows easy construction of
complex circuits without soldering [8].
The ultrasonic sensor functions on the same principles as
radar; it transmits a high-frequency signal and, based on the
echo, determines the proximity of a specific object.
Ultrasonic sensors can measure the distance of an object
accurately at a minimum of 2 centimeters and a maximum of
3 meters from the device. Therefore, we utilize ultrasonic
sensor to measure the distance between the user and
computer monitor. In our judgment, this sensor would be
enough to detect what we would consider a “bad posture”.
Two indicators were chosen for this design to notify the
user of bad posture, LED and Piezo speaker, in order to
make sure users get adequate alert to correct their bad
posture.
We did not provide an external battery, because
connecting Postuino to a USB cable is very convenient, since
the device is meant to correct bad posture of computer users.
Fig. 3 shows the circuit design of Postuino and Fig. 4 shows
the schematic for that circuit.
B. Distance Measurement
According to the United States Department of Labor, the
preferred viewing distance is between 20 and 40 inches (50
and 100 centimeters) from the eye to the front surface of the
computer screen, as shown in Fig. 5. To detect a bad posture,
we made the LED and Piezo speaker react when the distance
to the user falls below 20 inches (50 centimeters).
As sound travels at 1,130 feet per second, there are
73.746 micro-seconds per inch. This gives the distance
travelled by the ultrasonic sensor, outbound, and return, so to
find the distance of the user we take half of the distance
traveled.
distance in inches = (duration/74)/2 (1)
For metric system users, we use the following equation to
calculate the distance in centimeters, since the speed of
sound is 29 microseconds per centimeter.
distance in centimeters = (duration/29)/2 (2)
When the user leans too close to the computer screen, the
distance between the user, computer, and Postuino falls
bellows the defined threshold distance which triggers the
LED to light up and the speaker to play the melody. Once the
user corrects the posture, the distance goes beyond the
threshold which causes the LED to turn off and the speaker
to stop playing melody. The reason for this is because our
vision of “good posture” involves having the minimal
distance between the user and the computer, which implies
sitting straight without slouching.
C. Data Visualization
Postuino measures the time a user spends sitting and
using a computer. The measured time is divided into straight
time and slouch time. Since human brains process visual
information efficiently, we visualized the collected data
using Plotly.
Plotly is a platform for analyzing and visualizing data
streaming from any hardware device. Arduino API allows
continuously transmitting data or transmitting a single chunk
of data from Arduino and then making interactive graphs in
the browser.
We used Arduino StopWatch library to measure elapsed
time. Also, we used Plotly Arduino Ethernet Library to graph
the data. The graphs are embedded in Postuino’s web
application that will be described next.
D. Web Application
The accompanying web application was developed using
Bootstrap to allow easy switching between mobile, tablet,
and desktop views. Each user can sign up to create a new
account. Then, sign in to retrieve live feedback about the
time spent sitting either straight or slouching to use the
computer. The application also sends push notifications to
users who spent a certain amount sitting in order to take a
short break and purse working after the break. The default
time threshold is 20 minutes and it can be changed by the
user at any time using the web application.
The web application displays a two dimensional stacked
area chart for the straight sitting time and the slouching time
for the current day, five days, and one month. This chart can
be shared in social media channels or emailed to the user’s
physician or chiropractor. It also displays a two dimensional
bar chart to simplify the comparison between straight time
and slouching time. Fig. 6 displays both charts for a sample
user.
IV. EXPERIMENT DESIGN
In our study, we setup a workstation and mounted
Postuino on top of the screen to record the elapsed time,
including straight and slouch time. First, (N=24) subjects
used the computer for 3 consecutive hours after disabling the
alert system in Postuino in order to record the elapsed time,
without alerting the user to correct the bad posture. Then, we
repeated the first stage after enabling the alert system in
Postuino as follows; alert contains light only, sound only,
both light and sound. Finally, the second stage was repeated
after re-positioning Postuino to take place next to computer
monitor. We collected data with Postuino and a
questionnaire.
A. Participants
The 24 engineering graduate students recruited for the
experiment have experienced computer related injury during
their graduate study. All of them were required to sit and use
a computer for 3 consecutive hours in 3 different stages.
At the beginning of the experiment, all students were
asked questions regarding their gender, age, major, and type
of computer related injury they have. There were 9 females
and 15 males, all of age 24–35 as Table 1 shows.
Fig. 7 shows the distribution of our sample subjects by
major. Since all participants are enrolled in a Master’s degree
program, Fig. 8 shows the prevalence of computer use by
year of graduate study in first and second year. Finally, Fig.
9 shows the prevalence of muscle or joint pain among
engineering graduate students due to computer use.
B. Design Rationale
Several design choices were made during the experiment
design. Ultimately, we chose an experiment setup with the
following features: disabling Postuino alert system in phase 1
while using it to record elapsed time and enabling Postuino
alert system in phase 2 and 3. In the first part of phase 2 and
3, the alert system contains light only. In the second part of
phase 2 and 3, the alert system contains sound only. In the
third and final part of phase 2 and 3, the alert system
contains both light and sound. The difference between phase
2 and 3 is the position of Postuino. It is positioned on top of
the computer monitor in phase 2 and next to the monitor in
phase 3. There was 30 minutes break between the phases.
Two surveys were given to each participant. The first
survey is given to each participant before the experiment and
it contains questions about gender, age, major, hours per
week of computer use and the type of computer related
injury. The results of this survey are shown in Table 1 and
Fig. 7, 8, and 9. The second survey has 4 parts with the same
questions and a fifth different part. Part 1 will be answered
before the experiment; part 2 will be answered after phase 1;
part 3 will be answered after phase 2; and part 4 will be
answered after phase 3. Each part has a question about the
pain level during computing. The pain level is measured
using Wong-Baker faces paint rating scale as shown in
Fig.10. Finally, the fifth part has questions to review
Postuino’s efficiency.
C. Evaluation
According to our experiment, Postuino produced
significant effects in detecting and correcting bad posture.
17% of the participants felt less pain in phase 2. Of all
participants, 33% of the participants felt less pain in phase 3.
These numbers reflect the improvement in only one session.
Therefore, 87.5% of the participants agreed that Postuino can
improve their posture and 75% of the participants found the
break reminders in the web app are beneficial. Fig. 11, 12,
13, and 14 show the results of the second survey that was
answered by each participant after the experiment.
In regards to Postuino’s setup, 67% of the participants
preferred positing Postuino on top of the screen because it
gives better visibility. From our point of view, mounting
Postuino on top of the screen gave better readings. On the
other hand, 71% preferred light only, so they can use
Postuino in a public area such as library or laboratory.
Finally, Fig. 16 compares between straight time and
slouch time for a randomly selected user in the three phases
of the experiment. The minimum slouch time happened
when Postuino was mounted on top of the computer,
specifically in the third hour when both light and sound were
used as Figure 17 shows.
V. CONCLUSION
In our tests of Postuino’s efficiency, we have been able to
prove that improving the posture helps reducing muscle and
joint pain that is related to computer use. From quantitative
data, we can recommend the use of Postuino to detect bad
posture, minimize slouch time, and suggest frequent breaks
that minimizes the risk of getting computer related injuries.
Mounting Postuino on top of the screen gave better
readings. Still, users can position it next to the computer
monitor depending on their personal preference. Also,
enabling light alert only while disabling sound alerts allows
users to use Postuino in public places for maximum benefit
and better productivity. However, combining sound and light
alerts yields the best results in minimizing slouch time. In
sum, users have the option of choosing the appropriate alert
method depending on their needs and their situation.
TABLE I. SAMPLE DEMOGRAHICS
Participants
(N=24)
Gender: Male 62.5% (15)
Age: 24-29 75% (18)
Age: 30-35 25% (6)
Fig. 1. Postuino’s Side Position
Fig. 2. Postuino’s Front Position
Fig. 3. Postuino Circuit Design
Fig. 4. Postuino Schematic
20”
40”
Fig. 5. Preferred Viewing Distance
Fig. 6. Sitting Time Chart for a Sample User on the Web App
Fig. 7. Number of Students per Engineering Major
Fig. 8. Hours per week of Computer Use
Fig. 9. Prevalence of Computer-related Injurie
Fig. 10. Wong-Baker Faces Pain Rating Scale
Fig. 11. Pain Level During Computing (In General and During Phase 1, 2, 3)
Fig. 12. Postuino’s Preferred Position
Fig. 13. Postuino’s Preferred Alert Method
Fig. 14. Did Break Reminders help in improving the situation?
Fig. 15. Sraight Time vs. Slouch Time for a Randomly Selected User in Phase 1,2,
and 3
Fig. 16. Sraight Time vs. Slouch Time for a Randomly Selected User in Phase 1,2,
and 3
Fig. 17. Phase 2 breakdown ino 3 Hours
REFERENCES
[1] Charlotte Ord. 3 every day items that could be stopping you from
losing weight. November 7th, 2011.
http://charlotteord.blogspot.com/2011/11/3-every-day-items-that-are-
stopping-you.html
[2] United States Department of Labor Occupational Safety and Health
Administration
https://www.osha.gov/SLTC/etools/computerworkstations/component
s_monitors.html
[3] iPosture home page, http://www.iposture.com/
[4] Farella, E., Pieracci, A., Benini, L., & Acquaviva, A. (2006, June). A
wireless body area sensor network for posture detection. In
Computers and Communications, 2006. ISCC'06. Proceedings. 11th
IEEE Symposium on (pp. 454-459).
[5] Computer-related injuries (2014), www.betterhealth.vic.gov.au.
[6] Leah Buechley’s LilyPad Arduino. http://web.media.mit.edu/~leah/
LilyPad/index.html.
[7] Leah Buechley’s turn signal jacket, http://web.media.mit.edu/~leah/
LilyPad/build/turn_signal_jacket.html.
[8] Robbins, M., Johnson, I. P., & Cunliffe, C. (2009). Encouraging good
posture in school children using computers. Clinical Chiropractic,
12(1), 35-44.
... Paul P. Breen et al [4] worked on a biofeedback system with single accelerometer for real time correction of neck posture among computer users. Reem Alattas and Khaled Elleithy [9] from University of Bridgeport worked on a device, named Postuino, with the objective to warn computer users whether they are leaning too close to the computer screen. Rose Johnson and Janet van der Linden [10]'s study concludes that vibrotactile feedback can be used to notify the user of the device encourage the habit of changing their behavior. ...
... Paul P. Breen et al [4] work proved a biofeedback can bring significant improvement in posture correction. Reem and Khaled's [9] work reports over 50% of the graduating seniors at Harvard University experienced symptoms with computer use. Their work consists of an echo sensor placed on the monitor to detect distance and pose an alert. ...
... The design of the modular wearable robotic device is not user friendly and needs to be fitted with additional assistance, and absence of a companion application restricts the users from self monitoring [1]. Reem and Khaled [9]'s method of posture detection is only confined for those who work in front of computers. According to Rose Johnson and Janet van der Linden [10] iPosture is not able to correctly indicate which posture is correct, which might vary from person to person. ...
Thesis
Full-text available
This paper describes the development and design considerations behind the implementation of a cost effective smart posture trainer which will play a vital role to construct an ideal habit of posture in the day to day life for the users. The designed smart device detects the posture of the user based on the angle or on the sitting position of the user. The device is included with an android application which stores all the users’ data in different categories such as daily, weekly, monthly and yearly base in different views. It also provides multiple effective suggestions by learning the user data based on the condition or improvement of the posture of the users. So, in that case this smart posture assistance acts like a physician. In addition, the device, in its on state, measures the total number of footsteps passed by the user. Different features have been adopted to keep pace with modern technologies. The project uses modern engineering concepts to yield better implementation and better shape. Using the device in a user friendly manner with flexible portability and in a cost effective way to serve the majority is the primary goal. The aim of developing the smart posture assistance device also includes prevention for the non-affected users and curation to the users who are already affected with their back pain disease.
... One very innovative, application specific way for posture detection where the computer users are warned when they lean too close to the computer, is the "Postuino" [8]. ...
Preprint
Analysis of human posture has many applications in the field of sports and medical science including patient monitoring, lifestyle analysis, elderly care etc. Many of the works in this area have been based on computer vision techniques. These are limited in providing real-time solution. Thus, Internet of Things (IoT) based solution are being planned and used for the human posture recognition and detection. The data collected from sensors is then passed to machine learning or deep learning algorithms to find different patterns. In this chapter an introduction to IoT based posture detection is provided with an introduction to underlying sensor technology, which can help in selection for appropriate sensors for the posture detection.<br
... One very innovative, application specific way for posture detection where the computer users are warned when they lean too close to the computer, is the "Postuino" [8]. ...
Preprint
Full-text available
Analysis of human posture has many applications in the field of sports and medical science including patient monitoring, lifestyle analysis, elderly care etc. Many of the works in this area have been based on computer vision techniques. These are limited in providing real-time solution. Thus, Internet of Things (IoT) based solution are being planned and used for the human posture recognition and detection. The data collected from sensors is then passed to machine learning or deep learning algorithms to find different patterns. In this chapter an introduction to IoT based posture detection is provided with an introduction to underlying sensor technology, which can help in selection for appropriate sensors for the posture detection.
Article
This is the protocol for a Campbell review. The objectives are as follows: To investigate the evidence on the effectiveness of education programmes in improving the knowledge of back health, ergonomics and postural behaviour in University students.
Article
Full-text available
Objective Musculoskeletal problems reported by school children using computers have often been linked to bad posture. This study investigates whether posture education affects the reported prevalence of musculoskeletal symptoms amongst secondary school children using computers. Design A prospective blinded randomized controlled trial. Setting A school in Leicestershire, UK. Participants Seventy-one school children aged 11–12 years divided into intervention (n = 37) and control (n = 34) groups. Intervention Both groups received posture training delivered by teachers at the school and were assessed on their knowledge of correct posture. A follow-up lesson was delivered 1 week later during which the intervention group also received automated posture warnings and tips on their personal computers. Outcome measures The prevalence and severity of musculoskeletal symptoms were measured at the start of the study and at the start and end of the follow-up lesson and any differences between the two groups found over the course of the 60 min follow-up lesson noted. Results By the end of the follow-up lesson, the mean visual analogue pain scale representation of the degree of discomfort due to the musculoskeletal problems fell significantly from 1.53 to 0.39 for the intervention group, while that for the control group only fell from 1.23 to 1.13 (non-significant). The overall incidence of musculoskeletal problems in the intervention group showed a greater trend towards reduction, falling significantly from 32.4% to 5.4% compared with the control group, which fell from 29.4% to 20.59% (non-significant). Conclusions Postural interventions that include on-screen reminders during the course of the lesson significantly reduce the reported severity of discomfort of musculoskeletal problems and are associated with a trend towards lower reported frequencies of musculoskeletal problems overall. This data may be relevant to those devising ergonomic correction programmes for school children.
Conference Paper
Full-text available
The upcoming technology of Body Area Sensor Networks (BASN) has potential to enabling the design of Human Computer Interfaces (HCI). This type of applications poses new design challenges that are hard to be faced using traditional solutions. This paper present a novel solution for wireless and wearable posture recognition based on a custom-designed wireless body area sensor network, called WiMoCA. These nodes which are mounted on different parts of the human body make use of tri-axial accelerometers to detect body postures.
3 every day items that could be stopping you from losing weight
  • Charlotte Ord
Charlotte Ord. 3 every day items that could be stopping you from losing weight. November 7th, 2011.