Authors:
J. Angelin Jeba, S. Kaviya, V. Keerthana, C. Priyadharshini, S. Rubin Bose, M. Rehena Sulthana

Addresses:
Department of Electronics and Communication Engineering, S.A. Engineering College, Thiruverkadu, Chennai, Tamil Nadu, India. School of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India. School of Information Technology and Engineering, Melbourne Institute of Technology, Melbourne, Victoria, Australia.

Abstract:

Patients who have left the Intensive Care Unit (ICU) for the general wards have a high level of risk of experiencing unnoticed clinical deterioration since they are not subjected to round-the-clock monitoring. This paper presents the Sentinel Life Jacket, a smart wearable patient-monitoring device intended to provide real-time monitoring and risk assessment during the post-ICU recovery stage. The system combines the MAX30102 (heart rate and SpO2), MLX90614 (body temperature), and AD8232 (ECG and respiratory rate) sensors, along with an ESP32 microcontroller, to continuously monitor physiological parameters and transmit them wirelessly. The resulting data are also sent to a Flask-based server, where preprocessing methods, such as signal smoothing and noise reduction, are applied to improve signal stability. The machine learning model used is CatBoost to analyse vital parameters and categorise patient risk levels with high predictive accuracy. When abnormal patterns are detected, the system will send real-time alerts to health care professionals, update a live monitoring dashboard, and display important vitals locally on an OLED display. The suggested approach reduces response time and enhances patient safety during post-ICU recovery by continuously monitoring patients and detecting physiological anomalies early. The proposed hospital monitoring system is scalable, affordable, and reliable, using wearable sensing, IoT connectivity, and AI. Evaluation shows that the proposed system actively monitors patients and diagnoses physiological abnormalities, thereby reducing reaction time and improving patient safety after ICU discharge. 

Keywords: Post-ICU Monitoring; Wearable System; CatBoost Algorithm; Smoothing Technique; AI in Healthcare; Alert System; Intensive Care Unit; Monitoring Systems; Performance Analysis.

Received on: 26/05/2025, Revised on: 03/08/2025, Accepted on: 12/09/2025, Published on: 09/03/2026

DOI: 10.69888/FTSBE.2026.000670

FMDB Transactions on Sustainable Biomedical Engineering, 2026 Vol. 1 No. 1, Pages: 18-32

• Views : 38
• Downloads : 10

Download PDF