Telemonitoring in Emergency Care: Predictive Triage Models Using AI and IoT-Based Vital Sign Tracking

Abstract

Telemonitoring systems are increasingly becoming indispensable in emergency care settings due to their ability to provide continuous, real-time patient data. The integration of Artificial Intelligence (AI) with Internet of Things (IoT)-enabled vital sign monitoring has introduced a paradigm shift in predictive triage, enabling early detection of clinical deterioration and facilitating more accurate, timely decision-making. This paper investigates the development, design framework, and evaluation of AI-driven predictive triage models that utilize IoT-based sensors for monitoring vital signs such as heart rate, peripheral oxygen saturation (SpO₂), respiration rate, and blood pressure. The study explores the technical architecture, data acquisition process, machine learning model selection, and implementation challenges within emergency departments. A detailed case study demonstrates how predictive triage systems reduced response time, improved prioritization accuracy, and minimized preventable mortality in high-volume emergency settings. The paper concludes with policy recommendations, clinical implications, and future directions for scaling AI-enhanced telemonitoring solutions.