CORRELATION SYSTEM FOR PARKINSON’S

PATIENTS FOR GOOD HEALTH AND WELL-BEING

( SDG - 3 )

- SARVAJIT S S (22BCE1507)
 INTRODUCTION
• Parkinson’s disease (PD) is a chronic and progressive
neurodegenerative disorder that primarily affects movement control
but also has significant non-motor symptoms, such as speech and
cognitive impairments.
• This complex condition disrupts normal daily activities, as it
progressively hinders motor functions, leading to tremors, stiffness,
slow movements, and instability.
• According to the World Health Organization (WHO), Parkinson’s
affects over 10 million people globally, with increasing prevalence as
life expectancy rises. The Correlation System for Parkinson’s patients
is designed to address the challenges presented by this debilitating
disease through an innovative approach that motion monitors.
 MOTIVATION
• Parkinson’s disease (PD) is a progressive neurological disorder that
severely affects a patient’s motor and non-motor abilities, with
symptoms like tremors, stiffness, balance issues, and speech
degradation. These symptoms not only hinder basic mobility but also
limit the individual’s ability to communicate effectively, leading to a
decline in independence and quality of life.
• This separation limits the accuracy and responsiveness of such
systems, especially given the complex and fluctuating nature of PD
symptoms. For instance, a patient might experience posture
instability without concurrent vocal changes, or vice versa.
• This motivation highlights the urgent need for integrated,
affordable, and intelligent systems in managing chronic conditions
like Parkinson’s. Our project serves as a patient-centric
innovation that not only tracks symptoms but actively supports
 OBJECTIVES
• The Correlation System aims to assist Parkinson’s patients by
monitoring motion patterns, providing a proactive and adaptive
feedback mechanism. The system integrates for wearable motion
monitoring, bridging the gap between isolated symptom detection and
comprehensive, real-time monitoring.
• The system captures, filters, and analyzes signals using dedicated
microcontrollers and sensors, without the use of machine learning
algorithms.
• Real-Time Monitoring: Use a wearable belt with an MPU-6050 sensor to
detect deviations in posture.
• Develop a fully hardware-driven correlation system.
• Use sensors and microcontrollers to process input signals.
 LITERATURE REVIEW
Project Title Objective Solution Computin Parameter
Hardware
g s for
Drawbacks
Technolog Results
Components
y
Speech- To detect Non- Microphone, Machine Detection Not real-
Based and invasive Audio Learning accuracy time, lacks
Telemonitori monitor telemonitori interface (Regressio (>90%), integration
ng for Parkinson’ ng using n Models) symptom with
Parkinson’s s jitter, progression physical
(Tsanas et al., progressio shimmer, mapping movement
2010) n using HNR.
vocal
biomarkers
.
Wearable To assess Motion Acceleromete Signal Balance Sensor
Sensor Gait physical tracking r, Gyroscopes Processing deviation calibration
Monitoring activity with Algorithms metrics, complexity,
(Salarian et and gait acceleromet tremor lacks speech
al., 2007) instability ers and detection input
 LITERATURE REVIEW
Project Objective Solution Hardware Computing Parameter Drawbacks
Title Technology s for
Component Results
s

Dual-Task To examine Simultaneou Wearables + Training- Fall Lab setting

Systems in cognitive s cognitive- audio based reduction, only, no
PD and motor motor prompts performance symptom real-world
Therapy coordination feedback analysis response data
(Díaz et al., benefits of mechanisms time application
2021) dual-task .
training
Kinematic To detect Motion Wearable Pattern Fluctuation Lacks vocal
Monitoring motor fluctuation inertial recognition event count, symptom
with fluctuations tracking sensors algorithms gait freeze detection,
Motion using using detection limited
Sensors motion kinematic adaptability
(Rodríguez- sensors in algorithms.
 LITERATURE REVIEW
Project Objectives Solutions Hardware Computin Paramete Drawbacks
Title Components g rs for
technolog Results
y
Cloud- To propose Adaptive Mobile Cloud Real-time Data privacy
Based PD digital mobile- devices, cloud computing, alert issues,
Monitoring outcomes in based database adaptive response internet
Systems PD using feedback algorithms time, dependency
(Espay et integrated through symptom
al., 2016) mobile multiple tracking
health tools. modalities. frequency
Dual-Mode To explore Combined MPU-6050, Random Adaptive Prototype-
Adaptive multi-sensor acceleromet mic, vibration Forest, feedback level
Systems systems for er + speech motor embedded intensity, validation,
(Almeida et real-time PD input for ML response scalability
al., 2018) monitoring. real-time time, issues
feedback. usability
METHODOLOGY
 METHODOLOGY
HARDWARE CIRCUIT DIAGRAM :
 METHODOLOGY
PROPOSED ALGORITHM (PSEUDOCODE) :
BEGIN
INITIALIZE communication with Serial and MPU6050 sensor
IF MPU6050 connection fails THEN
PRINT error message
HALT execution
ENDIF
SET pinModes for:
- Vibration Motors (4 pins as OUTPUT)
- RGB LED (3 pins as OUTPUT)
- Buzzer (OUTPUT)
 METHODOLOGY
PROPOSED ALGORITHM (PSEUDOCODE) :
LOOP FOREVER
IF data is available on Serial THEN
READ incoming signal from Serial
CONVERT signal character (0-9) to integer value
ENDIF
READ accelerometer and gyroscope data from MPU6050
CALCULATE tilt angle using Y and Z axis (atan2 function)
IF tilt angle < 10 degrees THEN
SET RGB LED to GREEN
 METHODOLOGY
PROPOSED ALGORITHM (PSEUDOCODE) :
ELSE IF tilt angle between 10 and 20 degrees THEN
SET RGB LED to BLUE
ELSE
SET RGB LED to RED
FLAG high tilt as TRUE
ENDIF
IF tilt angle < 10 degrees THEN
CASE signal OF
0: SET LED to GREEN
1 or 2: SET LED to BLUE
>=3: SET LED to RED
 METHODOLOGY
PROPOSED ALGORITHM (PSEUDOCODE) :
ENDCASE
ENDIF
CASE signal OF
0:
SET tiltThreshold = 15
SET vibration intensity = LOW (100)
1 or 2:
SET tiltThreshold = 12
SET vibration intensity = MEDIUM (150)
>=3:
 METHODOLOGY
PROPOSED ALGORITHM (PSEUDOCODE) :
SET tiltThreshold = 10
SET vibration intensity = HIGH (255)
ENDCASE
IF tiltAngle exceeds tiltThreshold OR signal >= 1 THEN
ACTIVATE Vibration Motor 1 with calculated intensity
ELSE
TURN OFF Vibration Motor 1
ENDIF
IF signal >= 3 OR high tilt is TRUE THEN
TURN ON Buzzer
 METHODOLOGY
PROPOSED ALGORITHM (PSEUDOCODE) :
ELSE
TURN OFF Buzzer
ENDIF
DELAY 100 milliseconds
ENDLOOP

// Helper Function: setColor(red, green, blue)

// Controls the brightness of RGB LED using PWM
 HARDWARE REQUIREMENTS
COMPONENT FUNCTIONALITY SPECIFICATION IMAGES
Arduino Nano/Uno Microcontroller that reads - 16 MHz clock
sensor data and controls - 32 KB Flash memory
feedback devices - 14 digital I/O pins
- USB interface
MPU-6050 Sensor Detects patient posture by - 3-axis accelerometer
measuring acceleration + 3-axis gyroscope
and angular velocity - I2C interface
- ±2/4/8/16g range
Vibration Motor Provides haptic feedback - Voltage: 3V-5V-
(x4) when unsafe tilt or severity Vibration strength:
is detected Medium-High
depending on PWM
duty cycle
 HARDWARE REQUIREMENTS
COMPONENTS FUNCTIONALITY SPECIFICATION IMAGES
RGB LED Visual indication of - Three-color LED (Red,
symptom severity and Green, Blue)
tilt angle - Operates on 5V with PWM
control

Jumper Wires Electrical connections - Male-to-male/male-to-

between Arduino, female
sensors, and modules - 20–30 cm length

Breadboard/PCB Prototyping base to - 400-830 tie-points for

hold and wire prototyping
components together

Elastic Belt/Strap Mounting wearable - Adjustable,

system on the comfortable wearable
patient’s waist belt
 SOFTWARE REQUIREMENTS
SOFTWARE FUNCTIONALITY SPECIFICATION
Arduino IDE Write, compile, and upload - Supports C/C++
code to the Arduino - Compatible with all
microcontroller Arduino boards
MPU6050 Library Interfaces with the MPU- - Provides functions like
6050 sensor via I2C getMotion6() and sensor
calibration
Wire.h (I2C Library) Handles I2C - Built-in with Arduino IDE
communication between
Arduino and MPU-6050
Edge Deployment Software Runs software inference - Deployed on Arduino uno
and sends severity score or any Linux-capable
via Serial to Arduino system
 RESULTS AND ANALYSIS
PARAMETERS DESCRIPTION VALUES/SETTINGS

Testing Scenario Specific situations to be StablePosture , LowSeverity

simulated during testing Severe Speech Symptoms,
Stable Posture Postural
Instability,No Speech
Symptoms-Combined Speech
Symptoms and Posture Issues
Motion Input Conditions Different tilt angles and Safe Tilt Angle(≤15degrees)
movements tested Unsafe Tilt Angle(>15degrees)
Dynamic Movements
(Walking ,Sitting ,Turning)
Feedback Intensity Levels Levels of feedback based on Intensity(1-2 severity):Mild
severity scores vibration
Medium Intensity(3-4severity):
Moderate vibration
High Intensity(5severity):
Intensevibration
 RESULTS AND ANALYSIS
PARAMETERS DESCRIPTION VALUES/SETTINGS

Data Logging Frequency How often data is recorded Every Second

during tests On Significant Event(e.g.,
feedback activation)
User Feedback Collection Methods for collecting user Surveys(After Each Session)
feedback Observational
Notes(Throughout Testing)
Real Time Feedback(During
Testing)
Performance Metrics Metrics used to evaluate Detection Accuracy : 90-95%
system performance Response Time : ≤200ms
User Comfort Rating : 1-10
scale(Target : ≥7)
 RESULTS AND ANALYSIS
CONFUSION MATRIX : TRAINING CURVE :
 RESULTS AND ANALYSIS
Testing and Simulation : Test the system across a range of simulated
patient to evaluate its performance : Stable Posture with Low Severity
Symptoms Verifies if feedback remains passive and minimally intrusive
Data Logging for Performance Analysis: Enable data logging to capture
feedback events, tilt angles, and severity scores during testing. This
recorded data is analyzed to identify patterns, optimize tilt thresholds, and
adjust vibration levels. The insights gained help refine the correlation
algorithm to ensure precise, consistent feedback under varying conditions.
User Comfort and Usability Assessment : Collect feedback from test
users on comfort, ease of use, and effectiveness of the feedback. Adjust
component placement, vibration intensity, and LED brightness as needed to
balance effective feedback with user comfort. Iteratively improve the design
to ensure the device remains unobtrusive yet effective for continuous wear.
 CONCLUSION AND FUTURE
WORK
• The Correlation System represents a significant advancement in monitoring
and managing Parkinson’s disease symptoms by integrating motion
monitoring.
• The wearable device successfully monitors balance and posture in real-
time, providing immediate feedback to reduce fall risk, thus enhancing
patient safety and confidence in daily activities.
• Looking ahead, there are several avenues for further development and
enhancement. Future iterations of the system could expand sensor
integration by incorporating additional technologies, such as
electromyography (EMG) sensors and heart rate monitors, to provide a
more comprehensive view of patient health.
• Additionally, enhancing the user interface and experience would make the
wearable device more user-friendly, possibly introducing customizable alert
settings and mobile apps for caregivers to access patient data remotely.
 BUDGET
SERIAL NO. COMPONENT QUANTITY AMOUNT

1 ARDUINO BOARD 1 1900

2 MPU-6050 SENSOR 1 230

3 VIBRATOR MOTOR 4 60X4=240

4 LED 1 30

5 ELASTIC BELT 1 200

6 JUMPER WIRES AS PER 50

REQUIREMENT
7 BREAD BOARD 1 100

TOTAL AMOUNT : 2,750(APPROX)

 REFERENCES
• Tsanas, A., Little, M. A., McSharry, P. E., & Ramig, L. O. (2010). Accurate
telemonitoring of Parkinson’s disease progression by non invasive speech
tests. IEEE Transactions on Biomedical Engineering, 57(4), 884-893.
• Díaz, J. P., et al. (2021). Dual-task training for Parkinson’s disease: A
systematic review and meta-analysis. Neuroscience & Biobehavioral
Reviews, 120, 200-209.
• Almeida, Q. J., et al. (2018). Wearable devices for assessing gait and
balance in Parkinson’s disease: A systematic review. Journal of
NeuroEngineering and Rehabilitation, 15(1), 30.
• Pérez-López, C., Samà, A., & Rodríguez-Molinero, A. (2016). Signal
processing algorithms for detecting freezing of gait in Parkinson’s disease.
IEEE Transactions on Biomedical Engineering, 63(11), 2301-2308.