Architecture and Implementation of a patient criticality aware Edge-Cloud offloading technique

The Internet of Things (IoT)-based remote health monitoring is one of the most promising technological interventions that is emerging to address the unique challenges of affordability, accessibility, and availability in global health, which denote equitable access to healthcare, particularly in remote-rural-developing regions. IoT devices can monitor multiple vital signs, which are then used for AI-assisted decision-making systems, which in turn assist physicians in forecasting remotely. However, there are many difficulties, like bandwidth issues, data loss, and overburdened doctors due to the massive amount of data. Shifting data from cloud to edge improves performance, cost efficiency, privacy, reduces communication between distant servers and the edge, resulting in less processing delay. We developed a clinical response requirement based cloud-to-edge offloading technique. As a use case, we developed an edge AI application for acute hypotensive episode prediction and compared its performance both in the cloud as well as on the edge.