On the Synergy of IoMT Devices and Ceiling-Mounted Systems for Advanced Medical Data Analytics

The article explores a novel framework that combines Ceiling-Mounted Systems (CMS) and Internet of Medical Things (IoMT) devices to address critical challenges in healthcare data management. By leveraging the capabilities of IoMT devices for real-time data collection, the proposed CMS's robust sensing, storage, and processing features support efficient resource allocation in hospital environments. The proposed approach achieves approximately 35% reduction in latency, a 25% improvement in energy efficiency, and a 40% decrease in Age of Information (AoI) compared to traditional frameworks. The multi-objective optimization problem minimizes energy consumption and latency while ensuring fairness and timely data collection, which is particularly critical in patient monitoring and time-sensitive diagnostics scenarios. Deep Reinforcement Learning (DRL) methods solve the resource allocation challenge under realistic constraints. Specifically, Twin-Delayed Deep Deterministic Policy Gradients (TD3) and Soft Actor-Critic (SAC) algorithms are adopted to optimize task scheduling and system decisions in dynamic, resource-constrained settings. Simulation results demonstrate that SAC achieves approximately 20% faster convergence and 15% better adaptability in dynamic hospital environments compared to TD3, making it more suitable for real-time healthcare applications. These findings underscore the benefits of integrating IoMT devices with the proposed CMS infrastructures to meet healthcare requirements, such as robust security, high reliability, and real-time responsiveness.