Classification of Medical Big-Data collected using IoT Devices

Abstract

Modern medical systems manipulate a large number of clinical data collected using Internet of Things (IoT) devices. In this environment 5G network architectures provide low latency communication in order to support the strict constraints of real time medical services. Furthermore, the collected data need to be classified so their retrieval and manipulation to be immediate. This paper focuses on collecting big data from three types of sensors, namely body sensors, indoor and outdoor sensors. The collected data are stored in a Cloud infrastructure. A classification algorithm is proposed. For each type of sensor the algorithm classifies the received data into classes, to efficiently organize them into cluster of Virtual Machines (VMs). Specifically, taking into consideration the alternative solutions available in the literature, the proposed algorithm manipulates heterogeneous, large quantities of data and stores them in the cloud so that they can be retrieved in a short time, which is an important factor in cases where health data are involved.