Development of a Model for Health Monitoring Using Machine Learning and Data Mining System

Source abstract

Mobile Health is an emerging field that allows for real-time monitoring of individuals between routine clinical visits. It also makes it possible to remotely gather health signals, track disease progression and provide just-in-time interventions. Data Mining is the process of extracting valid, previously unknown, comprehensible and actionable information from large databases and using it to make crucial, critical business and logical decisions. The motivation is to improve on the existing by designing a medical bio-card, where the patient’s medical history can be accessed by any readily available medical personnel, other than the registered one (probably due to network issues or remote location), to proscribe drugs and commence treatment when the patient is in critical condition. The Objective of this research is to improve on the current system that focused on key risk factors such as High/ low Blood Pressure (H/LBP), Sugar level, Blood level, High Body Temperature (Fever), Tobacco use, Alcohol use, Inadequate Physical Inactivity, Unhealthy Diets, Abnormal Sleep Patterns etc which plays key roles in many chronic diseases. Hence, there is a need to continuously monitor at risk individuals for their health status and activities over extended periods of time in their natural settings with the goal of improving their health and well being using wearable sensors. Wearable devices are devices worn on, in or around the body, and it has highly reduced form factors, longer battery life and enhanced network capabilities to continuously and remotely monitor individuals in their natural settings over extended period of time. The Methodology employed for the study of the existing system and adopted for the design of the new system is Object-Oriented Analysis and Design Method (OOADM) since it is object-oriented in nature. The improved system implemented with an accelerometer, gyroscope and/or magnetometer will detect motion, orientation and direction. The mobile App developed using Python programming language, MySQLite and a Cloud Infrastructure (AWS). The result is in form of coupling these wearable devices with smart phones and cloud computing, where large volume of data can be remotely analyzed to find interesting patterns, detect abnormalities and detect target activities from continuous stream of data.