Intelligent Health Monitoring System

Abstract

This paper focuses on the hardware designs of motion sensor based microcontroller devices and auxiliary software that were built to enhance the practicality of device, in health monitoring and motion pattern recognition. The initiative is to build a low-cost device that can capture falling down pattern of an elderly and calls for help automatically, in response to the report of Center for Health Protection which enlisted falling down as 2nd deadliest killer last year. With the realization of this goal, chance of delayed rescue should be lowered. For hardware, three models were developed and each focused on different transmission method and sensor. This is to provide a full insight on one's motion pattern. The transmission methods implemented are Bluetooth & SD card storage, and sensors implemented are accelerometer & IMU (Inertial Measurement Unit). For software, terminals on PC & mobile device were built to receive data and perform real-time analysis. Graph plotter with filters is developed to plot output and have deeper analysis on motion pattern recorded. Machine learning on motion pattern is also covered.