Intelligent IoT and Big Data enabled HealthCare Monitoring System using Machine Learning Strategies

Abstract

In the Information Technology industry, the appliance of Internet of Things (IoT) and other associated technologies creates an illusion in development and standards. This technological development initiates several researchers to invent new things in various domains. The healthcare industry is the major concern of many researchers to do some innovations and provide an effective support to people in an innovative way. So, that a new technology is introduced on this paper called, Effective Machine Learning Strategy (EMLS), in which the proposed approach concentrates more on elder people as well as patients to continuously monitoring their health with respect to smart device. The smart device is comprised with lot of sensors to accumulate the patient related information and pass it to the monitoring unit. The sensors associated with the smart device are: Integrated Pressure Level Monitor Sensor, Body Temperature Estimation Sensor, MEMS Sensor and Temperature and Humidity Estimation Sensor. This all will be integrated with the smart device in association with Global Positioning System (GPS) to identify the present location of the respective patent. In this paper, the Integrated Pressure Level Monitor Sensor is used instead of using a heart rate estimation sensor. It estimates the blood pressure and the heartbeat level. The term IoT enables an active bridge between client and the server end, in which it is used to carry the sensor data to the server for processing. In server end, the open source scripting is enabled to capture the sensor data and process accordingly for sending alerts and take appropriate actions to inform the situation to the care taker/doctor. A machine learning strategy is applied to properly maintain the data robustness and the server efficiency, in which the detailed summary is provided over the methodologies section of this paper. The concept of BigData enables the data maintenance support to the proposed approach to handle the huge data in an innovative manner without any duplications.  The proposed approach proves the resulting efficiency on the resulting section with proper graphs and the outcome estimations