Implementation of Energy Efficient Fog based Health Monitoring and Emergency Admission Prediction System Using IoT

With rapid development in Information Communication Technology (ICT), Wearable Sensor Networks with Internet of Things (WSN-IoT) has produced several improvements in the smart world environment. One of the main research challenges in Wearable Sensor is energy, since all the sensor nodes operation depends on battery power consumption. Hence a new middleware has to be introduced between Wearable Sensor nodes and Cloud to reduce latency and Power Consumption problems. Overcrowding in hospital premise, detecting priority of hospital admission for patients, managing and monitoring health status of the patient constantly are daily problems in any health care system. Even though IoT based wearable sensors monitor health status of patients regularly and provide intent treatment in critical stage, but there is some block hole in that such as latency, energy issues and unawareness of medical execution plans and policies to preserve them from sudden attacks such as Heart attack. The proposed work is to implement energy efficient FoG based IoT network to monitor patients’ health conditions from chronic diseases and highlights utility of Deep Learning model for analyzing the health condition of patients and predicting Emergency readmission cases well in advance. This model is also compare with existing machine learning algorithms such as Gradient boosted, Decision tree, Random forest and Logistic regression to achieve more accuracy. This paper introduces preemptive interval scheduling algorithm with predictive analysis for constant monitoring of status for critical patients. By means of comparative analysis done in this work energy efficiency has been achieved prominently.