SMART AID DUSTER -IOT ENABLED FRAMEWORK FOR SMART BLACKBOARD CLEANING

Abstract

Abstract Currently, wearable devices are widely used to monitor physical conditions during exercise and daily activities, tracking parameters like steps, heart rate, and sleep patterns with advanced analytical features. Isokinetic training is extensively utilized for functional rehabilitation and evaluation, but excessive exercise can lead to muscle stress or injury. Effective exercise methods are crucial for both healthy and ill individuals. Muscle fatigue, a reduction in muscle strength or power, is often measured using the median frequency of the EMG power spectrum. Studies employ methods such as fast Fourier transform algorithms to calculate this frequency, with a decrease indicating the onset of fatigue. Additionally, spatiotemporal EMG signal analysis plays a significant role in recovery. We propose a system integrating a Mechanomyogram (MMG) detector to measure muscle fatigue in rugby players. This system provides real-time alerts on muscle fatigue and relaxation,. minimizing injury risks. The study includes an loT -based wearable device with an EMG sensor and Lab VIEW-based data acquisition for muscle activity imaging.

Specification

Field of Invention
In recent times, wearable technology has become an integral part of monitoring physical
health and performance, especially during exercise and day-to-day activities. These
devices, designed to track parameters such as step count, heart rate, and sleep patterns,
have advanced significantly, providing sophisticated analysis features. One critical
application is isok:inetic training, which is widely used in functional rehabilitation and
assessment. However, during such training, frequent dynamic contractions of the leg
muscles can lead to muscle strain or injury if not performed correctly. Therefore, learning
to exercise effectively is crucial for both healthy and sick individuals.
Muscle fatigue, characterized by a decrease in the muscle's maximum power or strength,
is often assessed as a fall in the muscle's force. This phenomenon isn't limited to primary
muscle exhaustion but is also influenced by various tasks. Muscle physiologists define
muscle fatigue as a severe exercise-induced decline in musck force. Fatigue is also
described as a reduced capacity for minimum performance, and the most effective method
to evaluate fatigue involves using the highest test of effort in an athlete's competitive
context.
Numerous studies have explored muscle fatigue monitoring through traditional methods
such as calculating the median frequency of the sEMG power spectrum. This power
spectrum is computed for signals using fast Fourier transform algorithms with tools like
MA TLAB. The median frequency for the recorded data is then calculated and plotted
against time, where a decrease in the mean value indicates the onset of fatigue. The median
frequency computation of the EMG signal is considered a primary and traditional indicator
of muscle fatigue status during intended muscle contractions. Spatiotemporal EMG signal
analysis is also employed to monitor muscle status, playing a crucial rule in recovery
programs.
To simplify and enhance the measurement of fatigue in athletes, we developed an
innovative system integrating Mechanomyogram (MMG) sensors to measure muscle
fatigue in rugby players during performance. This wearable monitoring system not only
alerts the user to the onset of fatigue but also provides information about muscle relaxation,
minimizing the risk of injury. The study is structured as follows:
Proposed architecture for implementing an loT-based wearable device with an
EMG sensor for monitoring
4 ·t .:i I ~ ·; ( ~ Q !. a i 4 I ;- .-, , ' ~ ' -·: .-:
Name Of the Applicant: Sri Sairam Institute of Technology, et al.
-Q)
C)
Ill
D..
Q)
-1- N
E...
0
-LL. -::1'
CIO
-::1'
"C"IO'"
..0.. ..
-::1'
-::1'
N
0
~
0
1.1') .....
(0
-.M... . -::1'
N
0
~
>
zQ :,. ·A. 'I' F... h,,i T'
I
.M... .
Lab VIEW-based data acquisition system for visualizing muscle. activity.
Background of Invention
The surge in mental health issues and the corresponding need for effective healthcare has
spurred research into the application of machine learning for mental health. This paper
offers a recent systematic review of machine learning techniques used to predict mental
health problems. In addition, it examines the challenges, limitations, and future directions
for applying machine learning in the mental health sector.
For this review, we collected relevant research articles and studies fr0111 reliable
databases, adhering to the PRISMA methodology. After screening and identifying
suitable studies, we included a total of 30 research articles. These articles are categorized
based on various mental health conditions, such as schizophrenia, bipolar disorder,
anxiety and depression, post-traumatic stress disorder, and mental health issues in
children.
In discussing our findings, we address the challenges and limitations
researchers face when applying machine learning to mental health problems. Moreover,
we provide concrete recommendations for future research and development in this field.
The rising average age of the population has resulted in an increased demand for
healthcare services. Advances in information and communication technology (ICT) have
paved the way for the development of smart cities, which include components such as
Smart Health
Hea)th aims to enhance healthcare by providing various services like patient monitoring
and early disease diagnosis. Nowadays, numerous machine learning techniques facilitate
these s-1-Iealth services.
f ! r ! I I'· [ !~ i-i (:: N N ,;, I 'i 4 ( 'i l { 'i ~7r ,, 4 i 4 ' ~ - ·,. ~ '' ' .. ~ ' ~ .,. ·,-: .,. ~ .:-:
-Q)
C),
Ill
D..
Q)
Naine Of the Applicant: Sri Sairam Institute of Technology; et al.
The results indicate that ML techniques are employed in many s-
1-!ealth applications, including the diagnosis of glaucoma, Alzheimer's disease,
bacteriolysis, ICU readmissions, and .cataract detection. The most commonly used ML
approaches are Artificial Neural NetworkS (ANN), Support Vector Machines (SVM), and
deep learning models, particularly Convolutional Neural Networks (CNN), which have
demonstrated high performance in most cases.
The advent of the Internet of Things (loT) has led to the development of smart health
monitoring systems, capable of tracking both mental and physical wellness. Key
contributors to various physical and mental disorders, such as stress, anxiety, and
hypertension, especially necessitate specific attention. Monitoring these conditions can
help detect probleins early, preventing long-term damage, improving quality of life, and
reducing caregiver stress and healthcare costs.
This study explores innovative technological solutions for real-time monitoring of stress,
anxiety, and blood pressure using discreet wearable sensors and machine learning
techniques. An automated artifact detection method was developed for blood pressure (BP)
and photoplethysmogram (PPG) signals, designed to automatically remove outlier points
. caused by movement artifacts from the BP signal. Eleven features extracted from the
oscillometric waveform envelope· were analyzed to determine the relationship between
systolic blood pressure (SBP) and diastolic blood pressure (DBP).
The proposed computational method for estimating blood pressure uses sophisticated
regression techniques to predict SBP and DBP values from PPG signal characteristics,
validating the effectiveness of this architecture.
Safer, better, simpler patient care: loT connectivity is revolutionizing healthcare, forging
innovative connections between patients and healthcare providers. Devices like wearables,
skin sensors, and home monitoring tools offer deeper medical insights into symptoms and
health trends, elevate remote care, and give patients greater control over their treatment.
Sensors play a key role in this ecosystem by enabling the real-time collection and analysis
of patient data. This allows healthcare providers to spend more time with patients and less
on logistical tasks.
( \ ' i I lc ;:: ~~ rt ~ N pj i\ I ·j 4 I ·; t { 'J i::r 2 .:i ·j 4
,. ' ~ ·- ,_. ::'.; '.:; ~
,. ~ ~ ~ -·: :-:
-Q)
C)
Ill
D..
Q)
Name Of the Applicant: Sri Sairam Institute ofTechnology, et al.
Semtech's LoRa wireless radio frequency technology, with its long range, low power, and
extended battery life, is perfect for loT-based healthcare applications that rely on batterypowered
sensors. Using unlicensed spectrum, these devices can communicate over public,
private, or hybrid LoRa WAN networks, both indoors and outdoors, complementing
existing technologies like Bluetooth, Wi-Fi, and Cellular.
In the rapidly evolving technological landscape, the outbreak and emergence of diseases
have become critical issues. Precaution, prevention, and disease control through
technology hi:we become major challenges for healthcare professionals and industries.
Maintaining a healthy lifestyle is increasingly difficult with busy work schedules. Smart
health monitoring systems offer solutions to these challenges.
The recent revolution of Industry 5.0 and SG technology has led to the development of
. smart, cost-effective sensors for real-time health monitoring. These Smart Health
Mon'itoring (SHM) systems provide fast, reliable, and affordable health services from
remote locations, which was not possible with traditional healthcare systems. Integrating
blbckchain frameworks has enhanced data security and privacy, protecting patients'
confidential data from misuse.
lnCOt:JJOrating deep learning and machine learning to analyze health data has achieved
preventive healthcare and better management of patient outcomes, facilitating the early
detection of chronic diseases. To further improve cost-effectiveness and real-time
capabilities, cloud computing and storage have been integrated.
Name of the Applicant: Sri Sairam Institute of Technology, et at.
-Q)
C)
Ill
D..
Q)
-1- N
E....
0
-LL. -::1'
CIO
-::1'
"C"IO'"
0 .....
-::1'
-::1'
N
0
~
0
Lt) .....
(0
-.M... . -::1'
N
0
~
>
~·ATCNT
I
.M... .
Summary
Currently, wearable devices are commonly used to monitor physical conditions during
exercise and daily activities, tracking metrics like steps, heart rate, and sleep patterns. A
new system is proposed that leverages loT and machine learning to detect muscle strain
and poor posture, providing alerts to users on how to correct them. Poor posture, such as
slouching over screens, can cause strain injuries and discomfort in various body regions.
To promote better posture, researchers have developed detection systems that alert users
when they need to adjust their posture. This new approach utilizes inertial measurement
unit sensors to detect slouching and electrical muscle stimulation to automatically correct
posture.
Objectives
The objectives of your proposed system for detecting muscle strain and poor
posture are:
• Monitor physical conditions during exercise and daily activities using wearable
devices.
• Detect muscle strain and incorrect postures using loT and machine learning
technologies.
• Alert users to correct their posture to prevent strain injuries.
• Promote good postural habits through slouch detection systems that notify users
when adjustments are needed.
• Implement inertial measurement unit sensors and electrical muscle stimulation to
correct posture automatically .
Brief Description of the Drawing
Fig 1 : Block Diagram for Muscle Strain Detection
The flowchart Fig. I illustrates The muscle strain detection and alert system operates
through a series of steps, beginning with data collection via wearable sensors that monitor
muscle activity and posture. This data is transmitted to a processing unit using loT.
Subsequently, the data undergoes preprocessing to remove noise and artifacts. Key features
related to muscle strain and posture are then extracted and analyzed using
1 4 /1 i ,. ·j 4 I ;-:-;: =-~
~ _,
"
Name of the Applicant: Sri Sairam Institute of Technology, et al.
-Q)
C)
Ill
D..
Q)
-1- N
E....
0
-LL. -::1'
CIO
-::1'
"C"I'O"
0 .....
-::1'
-::1'
N
0
~
0
Lt) .....
(0
M -..... -::1'
N
0
~
>
~'f\T~NT
I
M .....
machine learning algorithms. The system identifies instances of muscle strain and incorrect
posture, alerting the user with suggestions for correction. Additionally, inertial
measurement unit sensors and electrical muscle stimulation can automatically correct
posture. Continuous monitoring and feedback ensure proper posture maintenance and
muscle strain avoidance, with data stored securely for future reference and system
improvements.
Fig 2: Use case diagram of the proposed solution
The diagram in Fig.2 depicts Wearable sensors are used to monitor muscle activity and
posture. These sensors are typically placed on areas prone to strain or incorrect posture, like the
back, shoulders, and legs. They collect continuous data on muscle movements and body
alignment during various activities.
The collected sensor data is wirelessly transmitted to a processing unit using loT
technology. This ensures that the data can be analyzed in real-time, providing instant
feedback to the user.
Data Preprocessing: Once the data is received, it undergoes preprocessmg. This step
involves cleaning the data to remove any noise or artifacts caused by external factors such
as movement or electrical interference. This ensures the accuracy of the data used in
subsequent analyses.
·; 4
1 0. c, fl ~ 4 i 4 ( }, }, { I ;- c,
~ .,. -,...- _, _,
Name of the Applicant: Sri Sairam Institute of Technology, et al.
-Q)
C)
Ill
D..
Q)
-1- N
E...
0
-LL. -::1'
CIO
-::1'
"C"IO'"
0 ......
-::1'
-::1'
N
0
~
0
Lt) ......
(0
M -...... -::1'
N o.
~
>
:ll\T~NT
I .
M ......
Fig 3 : Circuit diagram for RFID connected with Arduino UNO
The image in Fig. 3 Key features related to muscle strain and posture are extracted from
the cleaned data. These features may include the duration and intensity of muscle
contractions, body angles, and movement patterns. Advanced algorithms identify these
critical indicators.
Machine Learning Analysis: The extracted features are then analyzed usmg machine
learning algorithms. These algorithms are trained on a vast dataset of muscle strain and
posture patterns, enabling them to accurately detect anomalies. The machine learning
model can classify the data to determine if the user is experiencing muscle strain or
maintaining poor posture.
Strain and Posture Detection: The system uses the machine learning model to identify
instances of muscle strain and incorrect posture. When such instances are detected, the
system logs them and prepares to alert the user.
User Alert: Upon detecting muscle strain or poor posture, the system sends an alert to the
user. This alert could be a vibration, sound, or visual notification on a connected device .
The alert also includes suggestions on how to correct the posture or relieve the muscle
strain.
Automatic Correction (Optional): For enhanced functionality, the system may include
inertial measurement unit (IMU) sensors and electrical muscle stimulation (EMS). IMU
sensors detect slouching or incorrect posture, and the EMS automatically stimulates the
muscles to correct the posture without the user's conscious effort.
Feedback Loop: Jhe system continually monitors the user, providing ongoing feedback. to
ensure they maintain proper posture and avoid muscle strain. This feedback loop helps
users develop better habits over time.
Data Storage and Review: All collected data is stored securely in the cloud or on local
•i Li
.1 1.
~- Q '.!. Li 1 4 .. .-, I ·t ~ ' I
~ ; ~· -' ,..:
-Q)
C)
Ill
D..
Q)
Name of the Applicant: Sri Sairam Institute of Technology, et al.
devices. This data can be reviewed later to track progress, analyze trends, and improve the
machine learning· model's accuracy. Data storage also ensures that the user's history is
available for future reference.
Output
In today's world, a significant portion of the population sutlers ti-om back pain, injuries,
neck pain, and shoulder problems, creating a growing need for effective solutions. To
address this, a wearable device has been designed. This device includes a mechanism to
detect areas of stress and the duration for which a person remains in the same posture.
Muscle fatigue, which indicates a weakening of muscle performance over time, typically
occurs during vigorous exercise. However, wearable devices . are now widely used by
individuals to monitor their physical activities. In response to this trend, an loT -enabled
wearable device has been proposed and implemented. This device, equipped with an
Arduino controller, EMG sensor, and WiFi module, monitors muscle fatigue in real-time.
This new loT-enabled wearable device, equipped with an Arduino controller, EMG sensor,
and WiFi module, is proposed to monitor muscle fatigue in real-time. The belt-shaped
wearable not only detects stress areas but also monitors the time spent in a particular posture.
For example, if a user sits for an extended period, the device can alert them to take a walk.
If the user adopts an improper sitting posture, the device will prompt them to correct it.
Additionally, the device can provide exercise, health, and mental well-being tips to the user.
-Q)
,C)
Ill
D..
Q)
-1- N
E....
0
-LL. -::1'
CIO
-::1'
"C"IO'"
0 .....
-::1'
-::1'
N
0
~
0
1.1') .....
(0
M
Name of the Applicant: Sri Sairam Institute ofTechnology, et at.
Detailed Description of the Invention
Sensors Muscle Flex Sensor (EMG) :
• Movement/tension in muscles will send an input signal which will be picked up by the
EMG
• Checks whether the muscle is tightened or loosened.
• Sends a high or low signal corresponding to the muscle tightened or loosened.
Accelerometer:
• Movement in the body, and changes in posture will change angles measured by the
accelerometer. Thus, body movement will. act as input to accelerometers.
• Measures the acceleration at the base of the curve of the spine and the top of the curve
of the spine. Thus, calculate the angle by which the person is bending and check if threshold
is crossed.
• These values are sent directly to the microcontroller which can be used to calculate the
angle differential.
Vibration Motor:
• Takes a high or low from the microcontroller
• Vibrates if it is high and indicates to the user that they are not in a position that has
acceptable posture.
• The output is the vibration and we will have a regulator to control the duration and
signature of vibration so that this is a comfortable signal for the person using this device
IMU:
• Data acquisition and angles should be accurate along x,y,z axis
• The IMU will be connected to' the microcontroller and rotated along the axis while
obtaining data in real time
Data Transmission Module:
Bluetooth Modem:
• Bluetooth modem should consume a low energy and should efficiently send data to a
user interface .
• Real time data sent to a user interface will be formed as a graph to allow the user view
:!:: his performance during a day.
-::1'
N
0
~
>
~.f\T~N·T QFf~-!~:~ (tt~Nt·+.fl.-.l
I
.M... .
Name of the Applicant: Sri Sairam Institute of Technology, et al.
-Q)
C)
Ill
D..
Q)
-1- N
E....
0
-LL. -::1'
CIO
-::1'
"C"IO'"
0 .....
-::1'
-::1'
N
0
~
0
Lt) .....
(0
-.M... . -::1'
N
0
~
>
~.A,T~~rT
I
.M... .
• A low energy Bluetoolh module will be used to avoid consuming a lot of energy.
Power Supply Module:
Solar Panels:
• Photovohaic cells for energy harvesting. The solar panels used in the Surface Buoy are
equipped with high-efficiency photovoltaic cells specifically designed for the marine
environment.
• These cells are chosen for their corrosion resistance and waterproofing capabilities. To
enhance durability and flexibility, amorphous silicon or thin-film solar panels may be
prefen·ed.
• These solar panels are strategically placed on the buoy's surface, ensuring optimal
exposure to sunlight throughout the day.
• To maximize energy capture, adjustable mounts ·or tracking systems may be
incorporated to follow the sun's movement
Battery Systems:
• Rechargeable batteries for energy storage.
• This stored energy can then be used during cloudy days or at night. Choose batteries that
are durable and can withstand harsh environmental conditions.
• The choice of rechargeable batteries, specifically selected for their suitability in harsh
marine environments, reflects a commitment to long-lasting and resilient energy storage.
• Regular monitoring and maintenance, including remote diagnostics, are incorporated to
identify and address any issues promptly, ensuring the consistent and reliable operation of
the Aquatic Ecosystem Monitoring System.
The idea under consideration is a complete system that utilizes technology to Improve
public transportation. It encompasses both commercial institution-based transport services
and metropolitan bus networks. To enhance user experience, operational efficiency, and
passenger safety, it combines machine learning, real-time data analytics, loT sensors, and
mobile applications. With capabilities like crowd recognition and real-time surveillance for
users of the transit system, the system, in its mature state, is also adaptable to individual
tracking for increased security within private institutions.
Application development Module:
• The application development module of the Muscle Strain Detection System is a
crucial component designed to provide a user-friendly interface for real-time monitoring,
analysis, and visualization of data collected from the deployed loT sensors .
I
Name of the Applicant: Sri Sairam Institute of Technology, et al.
-Q)
C)
Ill
D..
Q)
-1- N
E....
0
-LL. -::1'
CIO
-::1'
"C"I'O"
0 .....
-::1'
-::1'
N
0
~
0
Lt) .....
(0
M -..... -::1'
N
0
~
> 0,.··--- z'I'\!:\...NI
M .....
The application is developed with a focus on accessibility and ease of use, ensuring that
stakeholders, including environmental researchers, local communities, and government
agencies, can effortlessly engage with the system
Alert System Module:
• Alert system reads the position of the User data, analyze it and process the data and
provides the accurate result
• Using JOT and Machine Learning we can analyze the data of the User and provide a
Alert based system
• Example: If the User if sitting for a longer period, we can alert the User to go for a
walk and if the
User is not sitting properly then we can alert the User to correct the User
• We can also suggest the User with Exercise and Health and mental tip ..


We Claim,
Claim I: Real-time Monitoring: The project claims to provide real-time monitoring of
muscle fatigue and posture through wearable devices.
Claim 2: Advanced Technology Integration: By integrating loT, machine learning, EMG
sensors, and an Arduino controller, the system offers an advanced solution for detecting
and managing muscle strain and incorrect posture.
Claim 3: Comprehensive Detection: The system detects areas of muscle stress, monitors
time. spent in specific postures, and alerts users to take corrective actions.
Claim 4: User Alerts and Corrections: Users receive real-time alerts for poor posture and
muscle strain, along with suggestions for corrections and preventive measures.
Claim 5:Automated Posture Correction: With the inclusion of!MU sensors and electrical
muscle stimulation, the system can automatically correct posture without the user's
conscious effort.
Claim 6: Health Improvement: The system aims to improve overall health by encouraging
proper posture, reducing muscle fatigue, and preventing repetitive strain injuries.
Claim 7: Data Security and Privacy: By utilizing advanced data security measures like
blockchain, the system ensures the confidentiality and privacy of user data.
Claim 8:Cost-Effective and Accessible: The project claims that the wearable device IS
cost-effective and can be used by a wide range of individuals for health monitoring .
These claims highlight the comprehensive and innovative approach the project takes
towards enhancing physical health through modem technology. A system and method for
digital ticketing using mobile applications and smart cards, integrated with loT technology,
allowing passengers to purchase tickets for both public and private buses via a platform
developed using Flutter.

Documents