AI-based techniques on edge devices to optimize energy efficiency in m-Health applications

Abstract

The fast increase in the number of patients who need continuous monitoring by caregivers and the inequality between the number of patients compared with the number of doctors cause a burden for both doctors and patients. This one-to-one relationship poses a real scalability challenge in the healthcare systems. Resolving the problem by exploiting the fast developments in the fields of sensors, mobile phones, and wireless technologies to improve health systems is a critical approach. M-Health system accommodates the use of an edge device to send medical data over the wireless network toward the m-Health center to diagnose and control the case of the patient as fast as possible. However, the delivery of the substantial medical data is constrained by two factors, the wireless bandwidth provisioned from the network, as well as the energy consumption since edge devices limited to energy sources. As a result, implementing artificial intelligence (0) techniques at the edge devices before transmitting will enhance the overall energy efficiency of the m-Health system. Deep learning can be used on medical data to facilitate data exchange and summarization. This chapter will introduce mobile and smart health, edge computing, and different smart preprocessing techniques using AI and specifically deep neural networks to facilitate the transmission of the huge medical data from the edge devices while ensuring the optimization of energy efficiency.