Design and Development of a Painless, Non-Invasive Blood Sugar Monitoring System

Abstract

Diabetes mellitus (DM) is a chronic metabolic condition that requires regular monitoring of blood sugar to be managed properly. But the usual invasive methods for testing are painful and uncomfortable, which often makes people avoid doing it regularly. To tackle this issue, we’ve worked on creating a painless and non-invasive glucose monitoring device using Near-Infrared (NIR) Spectroscopy along with Photoplethysmography (PPG). Our system uses two NIR light sources with wavelengths at 940 nm and 1050 nm—these were selected because glucose absorbs light well at these points. A BPV10NF photodiode is used to detect the reflected light from the fingertip, and the signal is amplified using an OPA320 Transimpedance Amplifier. An ESP32 microcontroller handles signal processing. On top of this, we’ve added a machine learning model to better relate optical signals to actual glucose levels for improved accuracy. The system is linked to a Firebase-powered real-time mobile and web app, where users can check their sugar levels, trends, and history. We even integrated an AI-based health assistant (Gemini) to give personalized tips and support. Lab testing showed a clear relationship between how much light is absorbed and how much glucose is in the solution, proving that non-contact glucose estimation is actually possible. The device is compact, budget-friendly, and easy to use, which could make glucose tracking less of a hassle for patients. Looking ahead, we plan to shrink the device further and improve AI prediction to make it even smarter and more adaptable.

Introduction

Diabetes mellitus affects over 540 million adults globally, and numbers continue to rise. Regular glucose monitoring is essential, yet current methods—finger-prick blood tests—are painful, costly, and create medical waste. As a result, many patients avoid frequent testing, increasing the risk of serious complications. Non-invasive monitoring technologies such as NIR spectroscopy and PPG have shown promise, but existing devices remain expensive, bulky, or insufficiently accurate for daily use.

To address these challenges, the project proposes a non-invasive, affordable, real-time glucose monitoring system. It uses dual-wavelength NIR sensing (940 nm & 1050 nm) combined with machine learning, IoT connectivity, and a user-friendly dashboard. The goal is to make glucose monitoring painless, low-cost, and accessible for daily use.

Aim and Objectives

The system measures glucose optically based on light absorption through the skin, processed via:

• Carefully chosen NIR wavelengths

• A compact optical circuit (ESP32, photodiode, op-amp, MOSFET switching)

• ML models for accurate glucose estimation

• A Firebase-powered IoT dashboard

• Experimental validation using glucose-water solutions

The device integrates signal processing, ML, and cloud tools to provide real-time glucose tracking along with AI-based health feedback.

Literature Review Summary

Over two decades of research has shown:

• NIR and PPG are strong candidates for non-invasive glucose sensing due to glucose absorption behavior between 900–1700 nm.

• Improvements in sensor design and noise reduction have increased reliability.

• AI/ML models (regression, neural networks, SVM) significantly enhance prediction accuracy by handling signal variations due to skin type, temperature, and motion.

• Despite progress, commercially acceptable accuracy remains challenging, motivating further innovation.

Proposed Methodology

The system consists of three layers:

1. Sensing Layer:
   Dual NIR LEDs illuminate the fingertip; a photodiode detects reflected light; a TIA amplifies the signal.

2. Processing Layer:
   ESP32 handles ADC sampling, filtering, wavelength switching, and data transmission.

3. Cloud Layer:
   Firebase stores real-time data; a React dashboard visualizes glucose trends; an AI chatbot (Gemini) provides personalized tips.

The system operates using the Beer–Lambert Law, relating glucose concentration to light absorption.

Results and Discussion

Prototype testing confirmed:

• Stable sensing performance using dual wavelengths

• Clean signal conversion with the OPA320-based TIA (0.2–2.8 V output, <1 mV noise)

• Fast MOSFET switching (~12 μs) enabling accurate wavelength separation

• Photodiode response aligned with glucose concentration changes as expected

Conclusion

This research work presents the design and development of a non-invasive glucose monitoring system using Near-Infrared (NIR) Spectroscopy, along with Photoplethysmography (PPG), machine learning, and IoT-based real-time monitoring. The main aim was to create a low-cost, painless, and user-friendly method to estimate glucose levels without finger pricking or disposable strips. Based on lab testing, the system showed positive results and proved that it is possible to detect glucose concentration using light absorption. The prototype used two wavelengths — 940 nm and 1050 nm — that are known to interact well with glucose. As glucose levels increased, the output voltage decreased, following the Beer–Lambert Law. The setup, which included an ESP32 microcontroller, BPV10NF photodiode, and OPA320 amplifier, worked reliably. The system could process the signal and send it to Firebase, where it was displayed through a live dashboard built using ReactJS. A Gemini AI chatbot was also added to help users understand their glucose trends and get basic health suggestions. Experiments were done using glucose-distilled water samples with concentrations from 80 to 280 mg/dL. Both wavelengths gave accurate results, with 1050 nm performing slightly better. The circuit showed stable output, low noise, and consistent behavior during lab testing. However, there are still some limitations. The system has not been tested on human tissue, so real-world conditions like skin type, temperature, or pressure might affect accuracy. Also, calibration might change over time due to aging of components. These are important points to improve in the next stage. In future, the focus will be on reducing the size of the system to make it wearable (like a wristband or fingertip sensor). Clinical testing will also be done to compare it with standard glucometers. More improvements will include better optical design, smarter AI predictions, and optimizing power usage for longer battery life. To conclude, the project successfully showed that glucose levels can be estimated non-invasively using NIR light. While more work is needed for real-world use, this system provides a strong base for future development of wearable and smart health monitoring devices.

References

[1] Komal Lawand, Mahesh Parihar, Shital N. Patil, “ Design and development of infrared LED based non-invasive blood glucometer”, IEEE, New Delhi, India, 2015. [2] Aminah Hina and Wala Saadeh, “Noninvasive Blood Glucose Monitoring Systems Using Near-Infrared Technology A Review”, Higher Education Commission (HEC), Pakistan,2022. [3] Surya Prakash Maguluri, “A non-invasive glucometer”, 2022. [4] “Lancet study: More than 100 million people in India diabetic”, The Lancet, 2023. [5] Mr. Ravikumar Rajendraprasad, “A non-invasive glucometer”, India, 2020. [6] T. R. Khan, A. Mostofa and M. Dey, “Non-Invasive Blood Glucose Measurement Device: Performance analysis of Diffused absorbance method and Diffused Transmittance method using Near Infrared Light”, 2023. [7] M.-U.-S. Chowdhury, S. Roy, K. P. Aryal, H. Leung and R. Pandey, “FabSense - A DIY Approach for Development of Electrochemical Sweat Glucose Sensor”, IEEE 20th International Conference on Body Sensor Networks (BSN), Chicago, IL, USA, 2024. [8] M. S. Ali, N. J. Shoumy, S. Khatun, L. M. Kamarudin and V. Vijayasarveswari, “Non-invasive blood glucose measurement performance analysis through UWB imaging”, IEEE 3rd International Conference on Electronic Design (ICED), Phuket, Thailand,2016. [9] A. Hina, H. Nadeem and W. Saadeh, “A Single LED Photoplethysmography-Based Noninvasive Glucose Monitoring Prototype System”, IEEE ISCAS, Sapporo, Japan, 2019. [10] Zeynali, M., Alipour, K., Tarvirdizadeh, B. et al., “Non-invasive blood glucose monitoring using PPG signals with various deep learning models and implementation using TinyML”, Scientific Reports, 15, 581 (2025). [11] M. Naresh, V. S. Nagaraju, S. Kollem, J. Kumar, and S. Peddakrishna, “Non invasive glucose prediction and classification using NIR technology with machine learning”, Heliyon, vol. 10, no. 2, 2024. 38 [12] M. Tamilselvi and G. Ramkumar, “Non-invasive tracking and monitoring glucose content using near infrared spectroscopy”, IEEE International Conference on Imaging Systems and Techniques, India, 2015. [13] B. Gayathri, K. Sruthi, and K. A. Unnikrishna Menon, “Non-invasive blood glucose monitoring using near infrared spectroscopy”, IEEE Engineering in Medicine and Biology Society (EMBC), 2014. [14] V. P. Rachim and W.-Y. Chung, “Wearable-band type visible-near infrared optical biosensor for non-invasive blood glucose monitoring”, Sensors and Actuators B: Chemical, vol. 296, 2019. [15] Mohammad Goodarzi, “Selection of the most informative near infrared spectroscopy wavebands for continuous glucose monitoring in human serum”, Talanta, vol.146,2016. [16] M. Khalil, “Comparative Study of Different Measurement Sites Using NIR Based Non-invasive Glucose Measurement System”, Procedia Computer Science, vol. 83, pp. 987–992,2016. [17] Jyoti Yadav, Asha Rani, Vijander Singh, Bhaskar Mohan Murari, “Prospects and limitations of non-invasive blood glucose monitoring using near infrared spectroscopy”, Expert Review of Molecular Diagnostics, vol. 15, no. 2, pp. 277–289, 2015. [18] P. Zhang, D. Zhang, Y. Wang, X. Song, and A. G. Gregg, “Global healthcare expenditure on diabetes for 2010 and 2030”, Health Policy, vol. 101, no. 2, pp. 98–105, 2011. [19] Y. Heise, J. Tran, and B. Meier, “Recent advances in noninvasive glucose monitoring”, Medical Devices: Evidence and Research, vol. 4, pp. 45–54, 2011. [20] A. Tiwari, S. Pramanik, and S. Srivastava, “Non-Invasive Glucose Measurement Technologies: Recent Advancements and Future Challenges”, IEEE Sensors Journal, vol. 21, no. 2, pp. 1434–1445, 2021. [21] A. S. M. Z. Hossain, M. A. Hossain, and M. S. Islam, “A Non-Invasive Blood Glucose Monitoring Device”, IEEE RAAICON, 2020. 39 [22] M. A. Hossain, M. S. Islam, and A. S. M. Z. Hossain, “A Portable Non-Invasive Blood Glucose Monitoring Device”, IEEE ICEEICT, 2015 [23] ] A. K. Singh, P. K. Singh, and R. K. Singh, “Development and Evaluation of a Sensor-Based Non-Invasive Blood Glucose Monitoring System”, Proceedings, vol. 82, no.1,p.19,2023. [24] A. Sharma, B. Kumar, and S. Verma, “Non-Invasive IR-Based Measurement of Human Blood Glucose”, Proceedings, vol. 35, no. 1, p. 27, 2021. [25] S. V. S. R. K. Rao, K. S. R. Anjaneyulu, and P. V. N. R. Prasad, “Noninvasive Blood Glucose Monitoring Systems Using Near-Infrared Spectroscopy: A Review”, Sensors, vol. 22, no. 13, p. 4855, 2022. [26] Y. Zhang, Y. Wang, and C. Zhang, “Non-Invasive Blood Glucose Monitoring Technology: A Review”, Sensors, vol. 20, no. 23, p. 6925, 2020. [27] Rong Liu, Wenliang Chen, Yun Chen, and Kexin Xu, “Influence of temperature on the precision of noninvasive glucose sensing by near-infrared spectroscopy”, Proc. SPIE 6863, Optical Diagnostics and Sensing VIII, 2008. [28] Ping Zuo, Yingchun Li, Jie Ma, and SiLiang Ma, “Analysis of Noninvasive Measurement of Human Blood Glucose with ANN-NIR Spectroscopy”, IEEE ICNNB, Beijing, 2005. [29] Michael S. Boyne, David M. Silver, Joy Kaplan, Christopher D. Saudek, “Timing of Changes in Interstitial and Venous Blood Glucose Measured With a Continuous Subcutaneous Glucose Sensor”, Diabetes, Nov 2003. [30] Jason J. Burmeister, Mark A. Arnold, “Evaluation of Measurement Sites for Noninvasive Blood Glucose Sensing with Near-Infrared Transmission Spectroscopy”, Clinical Chemistry, vol. 45, no. 9, 1999. [31] Zanon, M., Sparacino, G., Facchinetti, A., Talary, M.S., Mueller, M., Caduff, A., Cobelli, C., “Non-Invasive Continuous Glucose Monitoring with Multi-Sensor Systems: A Monte Carlo-Based Methodology for Assessing Calibration Robustness”, Sensors, 2013. [32] Khadem, H., Nemat, H., Elliott, J., & Benaissa, M., “In Vitro Glucose Measurement from NIR and MIR Spectroscopy: Comprehensive Benchmark of Machine Learning and Filtering Chemometrics”, Heliyon, 2024.

Copyright

Copyright © 2025 Ishwari Patil, Payal Sahu, Amit Mahajan, Aditya Ajmire, Anshul Fakirde, Arman Shaikh. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.