An IOT Based Glucose and Blood Pressure Monitoring System with Machine Learning for Estimating Vital Signs and Abnormal Conditions

Abstract

In recent years, the integration of Internet of Things (IoT) technologies in healthcare has opened new possibilities for remote patient monitoring and early disease detection. Monitoring blood glucose levels is essential for managing and preventing conditions like diabetes. Conventional methods of glucose monitoring are typically invasive, expensive, and impractical for regular use. In this study, we present a smart, non-invasive system that utilizes Internet of Things (IoT) technology combined with machine learning algorithms to predict blood glucose status. The system measures physiological parameters such as body temperature, pulse rate, oxygen saturation (SpO?), and environmental humidity using DHT11 and MAX30100 sensors.These real-time readings are transmitted to the Blynk IoT platform using ESP8266 for continuous monitoring through a mobile application. Additionally, the collected data is fed into a Python-based machine learning model developed on Jupyter Notebook, where Gradient Boost Classifier and Random Forest Classifier algorithms are used to predict the patient\'s blood glucose status. The proposed system not only enables real-time health tracking but also assists in proactive healthcare management through predictive analytics. Experimental results demonstrate the effectiveness of the system in providing accurate monitoring and early insights into glucose level variations, making it a valuable tool for both personal and clinical health applications

Introduction

Overview

This project introduces an IoT-enabled health monitoring system integrated with machine learning (ML) to track vital health parameters and predict blood glucose status. It aims to provide real-time, remote, and cost-effective healthcare solutions, especially for managing chronic conditions like diabetes and hypertension.

System Capabilities

• Real-Time Monitoring of:

  • Body temperature

  • Humidity

  • Pulse rate

  • Blood oxygen saturation (SpO?)

• Predictive Analysis:

  • ML models estimate blood glucose status from sensor data.

  • Algorithms used: Random Forest Classifier and Gradient Boost Classifier.

Hardware Components

• DHT11 Sensor – Measures body temperature and environmental humidity.

• MAX30100 Sensor – Measures pulse rate and SpO?.

• ESP8266 (NodeMCU) – Microcontroller for data processing and wireless transmission.

Software Tools

• Arduino IDE – Programming the microcontroller.

• Blynk Platform – Mobile interface for real-time monitoring.

• Jupyter Notebook (Python) – ML model training and prediction.

Architecture & Methodology

1. Data Collection: Sensors collect real-time physiological data.

2. Data Transmission: Data is sent to the Blynk cloud via ESP8266 for real-time display on mobile.

3. Machine Learning Integration:

   • Sensor data is used to train ML models for glucose prediction.

   • Random Forest achieved 92% accuracy; Gradient Boost reached 89%.

Results

• High sensor reliability: Stable, accurate readings for all parameters.

• Smooth real-time monitoring: Effective transmission and user-friendly display via Blynk.

• Accurate predictions: ML models successfully inferred glucose status using non-invasive inputs.

• Low cost and portable: Easily deployable in home or rural healthcare settings.

Related Work

The study is positioned within ongoing research in IoT-healthcare, including:

• Fall detection systems

• Remote ECG and asthma monitoring

• IoT-based medication dispensers

• Early cancer detection and hydration monitoring

Future Improvements

• Incorporate more sensors (e.g., blood pressure, CGMs)

• Expand training datasets for better ML accuracy

• Enhance model performance and system scalability

Conclusion

In this work, an IoT-based health monitoring system has been successfully developed to continuously track vital parameters such as temperature, humidity, pulse rate, and SpO? levels. The integration of sensors with real-time data visualization through the Blynk platform enables easy and immediate access to critical health information for both patients and caregivers. Beyond simple monitoring, the system leverages machine learning models — specifically Gradient Boost Classifier and Random Forest Classifier — to predict the patient\'s blood glucose status, enhancing its capability as a proactive healthcare solution. The designed system emphasizes low-cost hardware, user-friendly mobile interfaces, and intelligent predictive analytics, making it highly suitable for both personal use and remote healthcare applications. By combining real-time monitoring with predictive modeling, the solution not only tracks current health status but also provides early warnings, enabling timely intervention and better management of chronic conditions. Overall, this project demonstrates how IoT and machine learning technologies can be effectively integrated to create smarter, more responsive healthcare systems. Future enhancements could include the addition of more physiological sensors, implementation of advanced deep learning models, and secure cloud-based data storage to further strengthen the system’s accuracy, scalability, and security.

References

[1] P. Sharma, R. Mehta, and S. Sahu, \"IoT-Based Fall Detection System for Elderly Care,\" International Journal of Advanced Computer Science and Applications, vol. 11, no. 12, pp. 368–374, 2020. [2] P. Sundaravadivel, E. Kougianos, S. Mohanty, and M. Ganapathiraju, \"Everything You Wanted to Know About Smart Health Care: Evaluating the Different Architectures, Models, and Applications of IoT for Health Monitoring,\" IEEE Consumer Electronics Magazine, vol. 7, no. 1, pp. 18–28, 2018. [3] O. Khalid, S. U. Khan, and N. Javaid, \"IoT-Based Asthma Monitoring System Using Environmental Sensors,\" International Journal of Distributed Sensor Networks, vol. 16, no. 7, 2020. [4] A. Jaiswal, R. Pandey, and S. Singh, \"Remote Blood Pressure Monitoring Using IoT,\" Proceedings of the 2021 6th International Conference on Communication and Electronics Systems (ICCES), pp. 971–975, 2021. [5] A. Mishra and P. Sinha, \"IoT-Based Smart Medication Reminder and Dispenser System,\" International Journal of Engineering Research and Technology (IJERT), vol. 9, no. 9, pp. 65–69, 2020. [6] S. Verma, V. Singh, and A. Kumar, \"IoT-Enabled Early Cancer Detection System Using Biosensors,\" Materials Today: Proceedings, vol. 49, pp. 4717–4722, 2022. [7] J. Sahoo and P. Pattnaik, \"IoT-Based Hydration Monitoring System for Athletes,\" IEEE Internet of Things Journal, vol. 9, no. 15, pp. 12794–12801, 2022.

Copyright

Copyright © 2025 Pavithra Devi M., Sivakamasundari P., Gobikasri D., Vigneshwaran K., Prakash M.. 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.