A noninvasive glucose monitoring system is a device that is used to monitor the glucose level in humon body without disturbing the cells and without pain.

1. NON-INVASIVE BLOOD GLUCOSE
MONITORING SYSTEM
Alsamad Ansari (54931) Rishabh Posti (54952) Ritik Goyal (54955)
Simmy (54956) Sonakshi Arya (54957)
Project-II (495-B)
Guided By – Dr. Paras

2. LIST OF CONTENTS
01
Introduction
0 2
Current
Challenges
03
Hardware and
Software
Requirements
06
Benefits
05
Implementation
04
How it
Works?
07
Future
Development
08
Conclusion

3. INTRODUCTION
• Diabetes, a common chronic disease affecting millions worldwide, requires
regular monitoring of blood glucose levels.
• However, the current invasive method, involving finger pricks, poses several
challenges such as pain, expense, and the potential spread of infectious
diseases.
• Moreover, long-term use of the invasive method can lead to tissue damage in
the fingers.

4. • By utilizing the variation in the intensity of NIR (Near-Infrared) light
received from the finger, we can accurately determine the glucose
level in the blood.
Near-Infrared Spectroscopy Range

5. Current Challenges
• Invasiveness and Discomfort
• Limited Continuous Monitoring
• Cost and Accessibility
• Usability and applicability
challenges
• Time Consuming

6. ARDUINO UNO (ATMEGA328P) LCD 16x2 DISPLAY
HARDWARE REQUIREMENTS

7. IR SENSOR MODULE PHOTODIODE (1450nm,1550nm)
IR LED (1450nm,1550nm) GLUCOMETER

8. Arduino IDE Python IDE (Google Colab)
SOFTWARE REQUIREMENTS

9. HOW IT WORKS?
• This project involves non-invasive monitoring of glucose using NIR
and pulse
sensor.
• Near-infrared spectroscopy is a method that uses the near-infrared
region of
electromagnetic spectrum.
• The basic pulse sensor consists of a light emitting diode and a
detector like a light detecting resistor or a photodiode.
• When a tissue is illuminated with the light source i.e. light emitted by
the led, the amount of light absorbed depends on the blood volume in
that tissue.
• The detector output is in the form of electrical signal and is
proportional to the pulse rate.

10. • We determine the blood glucose level by passing NIR radiation through a
region of the body we are interested in to monitor its glucose level.
• As light source, NIR LED from Thorlabs is used, with λ = 1300nm,
1450nm, and 1550nm.
• The correlation between absorbed radiation and glucose concentration id
determined by Beer Lambert Law.
• Photodiode voltage is proportional to near infrared light transmittance. It
is then correlated with blood glucose concentration.

11. BLOCK DIAGRAM

12. CIRCUIT DIAGRAM
Circuit Diagram Of Non-Invasive Blood Glucose Monitoring System

13. IMPLEMENTATION

14. DATA COLLECTION
• We conducted a study involving diabetic individuals to analyze the
relationship between their glucose levels and the corresponding
analog voltage readings.
• Glucose levels were measured using the invasive laboratory method
for all participants.
• Simultaneously, analog voltage readings were obtained using the
proposed hardware setup

15. CURVE FITTING
• To establish the relationship between glucose levels and analog
voltage, we employed polynomial regression analysis.
• We performed curve fitting by fitting polynomials of different orders
(1st to 5th) to the data.
• The evaluation helped us select the polynomial regression equation
that provided the best balance between accuracy and complexity.

16. • After evaluating the performance of the polynomial regression models
of different orders, we observed the following:
• The first-order polynomial (linear) had limited accuracy in capturing
the non-linear relationship between glucose levels and analog
voltage.
F(x) = 0.677*x -
107.11

17. • As the order of the polynomial increased, the models could capture
more complex relationships, potentially improving prediction accuracy.
F(x) = 665.091*x - 2.843*x*x + 0.005*x*x*x -
.0000044031*x*x*x*x + .0000000002407*x*x*x*x*x -
58324.419;

18. RESULT
• Root Mean Square Error = Sqrt((Sum of Square of Individual Value)/Total No. of Sets)
Root Mean Square Error = 22.80%

19. ARDUINO IMPLEMENTATION

20. BENEFITS
• Convenience and Comfort
• Improved Quality of Life
• Enhanced Compliance
• Real-Time Monitoring
• Early Detection of Highs and Lows
• Reduced Risk of Infections
• User Friendly Experience

21. FUTURE
DEVELOPMENT
• Artificial Intelligence Integration
• Wearable Technology
• Continuous Monitoring
• Enhanced Connectivity
• Predictive Analytics
• Personalized Recommendations
• Improved Sensor Technology
• Collaboration and Partnerships

22. CONCLUSION
• The research successfully demonstrated a strong relationship between the
sensor output voltage and glucose concentration through experiments.
• The proposed non-invasive glucose monitoring system showed good accuracy
and has low manufacturing and maintenance costs.
• The results of the prototype indicate a promising future for the implementation of
NIR technology in real-time and continuous non-invasive glucose monitoring.
• The proposed NIR spectroscopy experiment has great potential for non-invasive
continuous monitoring of glucose levels in the human body.
• Future studies will investigate the impact of variables such as skin roughness
and body fluid concentration to further improve calibration and system sensitivity.

23. REFRENCES
• https://www.e-dmj.org/journal/view.php?doi=10.4093/dmj.2019.0121
• https://www.researchgate.net/publication/323526055_Design_of_no n-
invasive_glucose_meter_using_near-infrared_technique
• https://iopscience.iop.org/article/10.1088/1742-6596/1361/1/012041
• Sherman M, “How do blood glucose meters work?,” Chem, vol. 13,
(2006).[Online].NewsAvailable:https://uwaterloo.ca/chem13news/sites/ca.
chem13news/files/uploads/files/343_dec_2006_pages_5-6.pdf.
• https://www.researchgate.net/publication/323526055_Design_of_no n-
invasive_glucose_meter_using_near-infrared_technique.
• https://chat.openai.com/?model=text-davinci-002-render-sha

24. THANK YOU