MedTwin AI: An IoT-Enabled Patient Digital Twin Framework for Personalized Health Monitoring and AI-Based Anomaly Detection

Author

Aryan Mali, Mashood Khan, Rudra Khairavle, Prof. Satish Kuchiwale

Abstract

 Remote patient monitoring can bring together wearable sensors, connected devices, and computational analysis to show health observations gathered outside clinics and hospitals. A stream of sensor readings, or an alert triggered by a threshold, is not by itself a patient digital twin, and a prototype that has not been validated must never be treated as a diagnostic device. This paper describes MedTwin AI, a proposed framework based on the Internet of Things (IoT) that connects timestamped physiological and motion observations to a structured virtual patient state, keeps recent history together with data-quality metadata, and supports review of anomalies through transparent baselines and candidate machine-learning methods. The planned prototype pairs an ESP32-class controller with a MAX30102 pulse-oximetry and heart-rate sensor and an MPU6050 inertial sensor, while further sensors remain optional and subject to verification. The study plan keeps offline analysis of public datasets distinct from later hardware testing. The candidate datasets, PhysioNet Wearable Device Dataset and PAMAP2, carry labels and populations that do not represent clinical emergencies. The paper sets out an architecture, a data schema, an evaluation protocol, implementation milestones, and privacy safeguards. At this stage no physical prototype, trained model, or experimental result is claimed. The intended contribution is a reproducible evaluation framework for testing whether a patient-specific or context-aware baseline offers value beyond fixed-rule monitoring on a clearly defined task.

Keywords

Digital Twin, Internet of Things, Remote Patient Monitoring, Anomaly Detection, Wearable Sensing, Personalized Baseline, Smart Healthcare

Full Text:

Download Paper PDF

References

Analog Devices. (n.d.). MAX30102: High-sensitivity pulse oximeter and heart-rate sensor for wearable health [Product documentation and datasheet]. https://www.analog.com/en/products/MAX30102.html

El-Warrak, L., & de Farias, C. M. (2024). The state of the art of digital twins in health—A quick review of the literature. Computers, 13(9), Article 228. https://doi.org/10.3390/computers13090228

Espressif Systems. (n.d.). ESP32 series datasheet. https://www.espressif.com/sites/default/files/documentation/esp32_datasheet_en.pdf

Hongn, A., Bosch, F., Prado, L. E., Ferrández, J. M., & Bonomini, M. P. (2025). Wearable physiological signals under acute stress and exercise conditions. Scientific Data, 12. https://doi.org/10.1038/s41597-025-04845-9 (Dataset version 1.0.1: https://doi.org/10.13026/he0v-tf17)

Katsoulakis, E., et al. (2024). Digital twins for health: A scoping review. npj Digital Medicine, 7, Article 77. https://doi.org/10.1038/s41746-024-01073-0

Liu, F. T., Ting, K. M., & Zhou, Z.-H. (2008). Isolation forest. In Proceedings of the 8th IEEE International Conference on Data Mining (pp. 413–422). https://doi.org/10.1109/ICDM.2008.17

Reiss, A. (2012). PAMAP2 physical activity monitoring. UCI Machine Learning Repository. https://doi.org/10.24432/C5NW2H

Shaik, T., Tao, X., Higgins, N., Li, L., Gururajan, R., Zhou, X., & Acharya, U. R. (2023). Remote patient monitoring using artificial intelligence: Current state, applications, and challenges. WIREs Data Mining and Knowledge Discovery, 13(2), e1485. https://doi.org/10.1002/widm.1485

Verma, N., Singh, S., & Prasad, D. (2022). A review on existing IoT architecture and communication protocols used in healthcare monitoring system. Journal of The Institution of Engineers (India): Series B, 103, 245–257. https://doi.org/10.1007/s40031-021-00632-3

Share your valuable work from Social Media Buttons