Development of android based health monitoring system
Loading...
Date
2018-06
Authors
Ooi, Yoong Khang
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The large number of population within the age between 25 to 29 implied the
importance of health care services to maintain their wellbeing. However, a lot of health
measuring devices are only able to measure one health parameter. Besides, most of the
measuring devices stored only a limited number of records. Furthermore, these measuring
devices do not predict user’s health condition after health data is obtained. Therefore, this
project aims to develop an improved health monitoring system. The system consists of
multiple sensors. The system utilized heart rate click module to capture heart rate and
oxygen saturation level, and TMP007 sensor to obtain body temperature. These sensors
are integrated into Arduino microcontroller for data acquisition. The data will then be
sent to Raspberry Pi 3 via serial communication. The data read by Raspberry Pi 3 will be
used for health condition prediction through Support Vector Machine (SVM)
classification model. After that, all the health informatic data and health condition will be
stored in Microsoft Azure Cloud database. All the health parameter values and health
condition are stored together in a table of six columns under Cloud database using
MySQL query command. The mobile apps can be retrieved all data from Cloud database
using MySQL query command as well correspond to user name. To add security to the
stored health data, this project has developed an Android based app having face
recognition login system for the users to view their health data. The SVM classification
model achieved an overall accuracy of 93.33% from 60 testing data meanwhile FaceNet
classification model for face recognition achieved an overall accuracy of 99.0%. Both
classification models are accepted to be used for this project.