Novel Data Analysis Framework in Smart Healthcare

Abstract

Managing and processing the massive amount of heterogeneous data collected from IoT is a difficult task, and today’s infrastructure resources have been confronted as a result. Moreover, it becomes difficult to provide the user with significant information to make a decision due to the existence of massive information records. It is now possible to extract critical insights from massive streams of information in real time using complex event-processing (CEP) technology. Here an attempt is made to design a system that can generate complex events from large sensor data and send only those events to the cloud server. For generating complex events, the sensor’s threshold values are taken into consideration. If the sensor reading surpasses the threshold boundary, an event is formed and the associated data is forwarded to the cloud (in our case, we are using a temperature sensor, a pulse oximeter, and an ECG sensor). It avoids sending unnecessary data on the cloud and, subsequently, reduces data transfer time and memory space. Esper library is used for the CEP approach. The Viterbi algorithm and the support vector machine (SVM) are used to estimate the patient’s condition by delivering an emergency notice is to the physicians.