An IoT-Based Heart Rate and SpO₂ Wearable Device for Centralized Patient Monitoring in Viet Nam: Design and Preliminary Assessment
Main Article Content
Abstract
This paper presents the design and experimental prototyping of a wristwatch-style smart wearable device based on Internet of Thing (IoT) architecture for continuous centralized monitoring of heart rate and blood oxygen saturation (SpO₂). To address the critical limitations of prior works that remain constrained to isolated, individual-scale trials, the proposed solution establishes an optimized network topology that enables a centralized server to concurrently and synchronously manage high-throughput data streams from multiple distributed nodes. The hardware architecture optimizes microcircuit real estate on a 2-layer printed circuit board (PCB) platform with a compact dimension of 40 x 30 mm, integrating a high-performance dual-core ESP32-S3 microcontroller, a MAX30102 optical biosensing module, a DS3231 real-time clock (RTC), and an ST7789 TFT display interface. To effectively eliminate wrist-specific motion artifacts, an on-chip Adaptive Edge-based Kalman Filter (AE-KF) algorithm is implemented directly to preserve the underlying physiological morphology of the photoplethysmography (PPG) waveforms. The preprocessed time-series data are encapsulated into a JSON payload structure and transmitted in real-time to a centralized dashboard via the lightweight Message Queuing Telemetry Transport (MQTT) protocol over a Wi-Fi wireless network. The system was experimentally evaluated on a sample cohort of 100 elderly volunteers (aged 50 to 80 years) under both sedentary resting states and light physical activities, cross-referenced directly against a clinically validated LK87 fingertip pulse oximeter as the baseline gold standard. The empirical results confirm the superior accuracy of the proposed system: the root-mean-square error (RMSE) for heart rate monitoring is 1.5229 bpm with a strict Bland-Altman limit of agreement within ±3 bpm, and the SpO₂ measurement RMSE remains stable below 0.65. This study establishes a standardized data infrastructure ready for advanced machine learning algorithms to enable early clinical risk prognosis and remote preventive healthcare.
Keywords
IoT, ESP32-S3, SpO2, PPG, Heart rate
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
References
[2] R. Uddin and I. Koo, “Real-Time Remote Patient Monitoring: A Review of Biosensors Integrated with Multi-Hop IoT Systems via Cloud Connectivity,” Applied Sciences, vol. 14, no. 5, art. no. 1876, 2024, doi: 10.3390/app14051876.
[3] J. Allen, “Photoplethysmography and its application in clinical physiological measurement,” Physiological Measurement, vol. 28, no. 3, pp. R1–R39, 2007, doi: 10.1088/0967-3334/28/3/R01.
[4] K. B. Kim and H. J. Baek, “Photoplethysmography in Wearable Devices: A Comprehensive Review of Technological Advances, Current Challenges, and Future Directions,” Electronics, vol. 12, no. 13, art. 2923, 2023, doi: 10.3390/electronics12132923.
[5] D. Lapsa, R. Janeliukstis, M. Metshein, and L. Selavo, “PPG and Bioimpedance-Based Wearable Applications in Heart Rate Monitoring—A Comprehensive Review,” Applied Sciences, vol. 14, no. 17, art. 7451, 2024, doi: 10.3390/app14177451.
[6] W.-K. Beh, Y.-C. Yang, and A.-Y. Wu, “Quality-aware signal processing mechanism of PPG signal for long-term heart rate monitoring,” Sensors, vol. 24, no. 12, art. no. 3901, 2024, doi: 10.3390/s24123901.
[7] P. H. Charlton, V. Marozas, E. Mejía-Mejía, P. A. Kyriacou, and J. Mant, “Determinants of photoplethysmography signal quality at the wrist,” PLOS Digital Health, vol. 4, no. 6, e0000585, 2025, doi: 10.1371/journal.pdig.0000585.
[8] R. Kirszenblat and P. Edouard, “Validation of the Withings ScanWatch as a wrist-worn reflective pulse oximeter: Prospective interventional clinical study,” J. Med. Internet Res., vol. 23, no. 4, e27503, 2021, doi: 10.2196/27503.
[9] K. D. Torp, P. Modi, E. J. Pollard, and L. V. Simon, “Pulse oximetry,” in StatPearls [Internet], Treasure Island (FL): StatPearls Publishing, updated Jul. 30, 2023.
[10] N. Fong, M. S. Lipnick, E. Behnke, et al., “Open Access Dataset and Common Data Model for Pulse Oximeter Performance Data,” Sci. Data, vol. 12, art. no. 570, 2025, doi: 10.1038/s41597-025-04870-8
[11] M. Cao, R. Ramezani, V. K. Katakwar, W. Zhang, D. Boda, M. Wani, and A. Naeim, “Developing remote patient monitoring infrastructure using commercially available cloud platforms,” Frontiers in Digital Health, vol. 6, art. no. 1399461, 2024, doi: 10.3389/fdgth.2024.1399461.
[12] P. K. Sharma and R. K. Dhillon, “An Evaluation of MQTT and CoAP Protocols for Centralized Remote Patient Monitoring in Hospital Environments,” Computer Networks, vol. 243, art. no. 110285, 2024, doi: 10.1016/j.comnet.2024.110285.