Energy-efficient edge based real-time healthcare support system

The ubiquitous usage of wearable IoT (wIoT) devices has created a formidable opportunity for remote health monitoring system to provide paramount services such as preventive care and early intervention for populations at risk. The cloud-edge paradigm can efficiently manifest the complex computations required in providing these services. But the challenge in its exertion lies in incorporating intelligence at the edge devices. With the deluge of data availability, deep learning methods are very promising to obtain sufficient performance in healthcare applications. As the edge devices are resourceconstrained in terms of compute capability and energy consumption, unleashing deep learning services from the cloud to the edge requires efficient tackling of the exorbitant computational and energy requirements of deep learning frameworks. In this chapter, an energy-efficient smart edge based health care support system (EESE-HSS) is proposed for diabetic patients with cardiovascular disease. The proposed cloud-edge paradigm makes use of a hierarchical computing architecture for exerting expeditious diagnosis during emergencies. The intelligence framework incorporated at the edge is also built in an energy-efficient manner. Thus, the proposed healthcare support system has better efficacy in terms of energy efficiency and reduced latency. This makes it very supportive for fall detection in diabetic patients with cardiovascular disease who are susceptible to the risk of heart attack, stroke, heart failure and other vicious diseases.