Digital Twin for Battery-Powered Medical IoT Fleet: Safety-Aware Energy Budgeting Across Sensors, Edge AI, and Data Transmission

The rapid adoption of Medical Internet of Things (mIoT) fleets for remote patient monitoring is consistently hampered by the energy-safety paradox: aggressive power saving often risks the latency and reliability of life-critical alerts. In this paper we propose a Digital Twin-enabled Energy Budgeting Framework (DTEBF) for managing a heterogeneous fleet of battery-powered medical sensors. This hybrid Battery Digital Twin (DT), which merges the best of physics-based electrochemical modeling with data-driven predictors has been integrated into a framework for real-time estimates on State-of-Health (SoH) and Remaining Useful Life (RUL). We propose a new Safety-Aware Energy Budgeting (SAEB) algorithm to wisely allocate energy in an adaptive manner across three essential layers, including onboard sensing, Edge AI inference and wireless transmission based on the Clinical Urgency Score (CUS). Experimental results on the NASA Battery Dataset and simulated medical workloads demonstrate DTEBF achieves 39.4 % greater operational uptime and 73.3 % lower critical alert latency than traditional static power management approaches. Bridging the gap between hardware-centric energy optimization and clinical-grade reliability, this work enables scalable resilient long-term mIoT deployments.