The proposed system is an advanced integration of Internet of Things (IoT) and wearable technology, designed to enablereal-time,non-invasivemonitoringofbothstresslevelsand bloodglucoseconcentrationinhumans.Itcombinesphysiological sensing, embedded processing, and cloud connectivity to deliver continuous health insights. The stress detection module utilizes a heart rate sensor and a Galvanic Skin Response (GSR) sensor to monitor changes in pulse rate and skin conductivity. For glucose estimation, Near-Infrared (NIR) spectroscopy is employed. An ESP32 microcontroller serves as the system’s core, collecting sensor data, preprocessing it, and wirelessly transmitting it to cloud-based dashboards. A Python-based simulation environ- ment facilitates debugging and analysis. The system offers a compact and scalable solution for wearable applications. Future enhancementsincludemachinelearningintegrationforimproved calibration and personalization
Introduction
Overview
The system addresses the growing need for continuous, non-invasive health monitoring, particularly for psychological stress and blood glucose levels—both of which are linked to chronic conditions like hypertension, diabetes, and cardiovascular disease. Traditional methods involve clinical tools and invasive sampling (e.g., finger pricks), which are not ideal for frequent monitoring.
This project leverages IoT and wearable technologies to enable real-time, remote, and user-friendly health tracking.
System Architecture
Central Unit:
ESP32 microcontroller handles sensor input, processing, and wireless communication.
Key Hardware Components:
MAX30102 Heart Rate Sensor: Measures pulse via photoplethysmography.
GSR Sensor: Assesses skin conductance to infer stress.
Optional: OLED display for vitals, buzzer for abnormal readings.
Software Architecture:
Captures and digitizes sensor data.
Processes stress and glucose estimates.
Sends results to cloud platforms like Firebase or ThingSpeak via Wi-Fi.
Methodology
Sensor Data Collection:
Heart Rate: Detected using peak analysis.
GSR: Resistance derived via voltage divider circuit.
NIR: Reflectance levels mapped to glucose concentration.
Signal Processing:
Data cleaned and analyzed on-board.
Custom algorithms compute stress index and estimate glucose levels.
Simulation and Results
Environment:
Simulated in Python using libraries like NumPy and Matplotlib.
Generated Data:
Heart Rate: 65–105 BPM
GSR: 8–30 k?
NIR readings: 600–2800 units
Outputs:
Graphs for heart rate, GSR, stress index, and glucose.
Alerts triggered when thresholds are exceeded.
Cloud Integration:
Real-time data successfully uploaded to cloud dashboards.
Performance:
Accurate stress classification using HR and GSR.
Reliable glucose estimates from NIR mapping.
Successful wireless data transmission.
Conclusion
Thesystemintegratesnon-invasivestressandglucosemonitoringusingIoT.Itfeatureswearabledesign,cloudconnectiv- ity,andPython-basedsimulationforvalidation.Thecombined use of MAX30102, GSR, and NIR with ESP32 provides a reliable, scalable health monitoring platform.
References
[1] G. Picard, A. Turner, and R. Foster, “Stress detection using wearablephysiological and sociometric sensors,” Computer, vol. 48, no. 10, pp.26–33, 2015.
[2] S. A. Inamdar and S. S. Kulkarni, “Stress detection and monitoring using wearable biosensors,” in Proc. IEEE ICCUBEA, 2017.
[3] D. A. Hall and A. T. Bartholomeusz, “Non-invasive glucose sensing using near-infrared spectroscopy,” J. Diabetes Sci. Technol., vol. 5, no. 1, pp. 62–69, 2011.
[4] R. Palaniappan, “Non-invasive monitoring of blood glucose using sen- sors and AI,” Sensors and Materials, vol. 32, no. 7, pp. 2545–2556, 2020.
[5] C. Setz et al., “Discriminating stress from cognitive load using a wearable EDA device,” IEEE Trans. Inf. Technol. Biomed., vol. 14, no. 2, pp. 410–417, 2010.
[6] M. Gjoreski et al., “Continuous stress detection using a wrist device: In laboratory and real life,” in Proc. ACM UbiComp, 2016, pp. 1185–1193.
[7] J. A. Healey and R. W. Picard, “Detecting stress during real-world driving tasks using physiological sensors,” IEEE Trans. Intell. Transp. Syst., vol. 6, no. 2, pp. 156–166, 2005.
[8] J. Yadav et al., “Prospects and limitations of non-invasive blood glucose monitoring using near-infrared spectroscopy,” Biomed. Signal Process. Control, vol. 18, pp. 214–227, 2015.
[9] S. K. Vashist, “Non-invasive glucose monitoring technology in diabetes management: A review,” Anal. Chim. Acta, vol. 750, pp. 16–27, 2013.
[10] N. S. Oliver et al., “Glucose sensors: A review of current and emerging technology,” Diabet. Med., vol. 26, no. 3, pp. 197–210, 2009.
[11] S. M. R. Islam et al., “The Internet of Things for health care: A comprehensive survey,” IEEE Access, vol. 3, pp. 678–708, 2015.
[12] F. Mehmood and M. Ahmad, “A smart IoT-based system for monitoring diabetic patients,” Int. J. Adv. Comput. Sci. Appl., vol. 11, no. 6, pp. 1–7, 2020.