A dynamic power-aware strategy for Smart Health applications

The rapid proliferation of sensors, mobile devices, wireless connectivity and data availability, have influenced medicine significantly. There is a strong bond between Internet of Things (IoT) and Artificial Intelligence (AI) to improve the quality of healthcare, turning traditional into smart healthcare. Their convergence offers great opportunities, can improve patient outcomes, reduce medical errors, lower costs, overcome the existing medical systems limitations and avoid the degeneration of harmful situations. Despite these potential benefits, these new opportunities introduce some troubles. The focus of the paper is power management of IoT devices, as they are battery-operated and energy saving become one of the biggest concerns in this field. We propose an architecture enriched by a power-aware strategy, that taking into account the energy availability of the devices, allows to collect and analyze huge volumes of heterogeneous clinical data in order to monitor health status of patients and improve clinical decision. The proposed strategy consists of two main elements, a distributed service to logically sort IoT devices, and an allocation algorithm for distributing the computational load among that devices. The experimental results show the effectiveness of the approach able to achieve energy savings by means of the proposed power-aware strategy.