Grid-Based Prediction model for Coronary Heart Disease: Using Data Generated from the IoT-based Health Monitoring Systems
Ogungbade BA

Keywords

internet of things
coronary artery disease
radio frequency identification
health monitoring system

Abstract

Abstract

Internet applications are now closely associated with human life and hence, have become inevitable in human daily activities. This has resulted in the deployment of various types of sensors and computing devices in huge numbers. One of such is the network of objects which are interconnected to create and share data, called the Internet of Things (IoT). Amongst the applications enabled by the Internet of Things, the continuous health monitoring system (HMS) is a particularly important one. Therefore, this study was carried out to design architecture on fog network, that acquires data from IoT-based Health monitoring systems using Radio Frequency Identification-RFID, and develop a prediction model that predicts coronary artery disease using Artificial Neural Network. The existing heart disease datasets from the UCI Machine Learning Repository was used. In our result, the confusion matrix, in the order of (True Positive, False Positive, True Negative, False Negative) gives 41%, 8%, 40.3%, 10.6% respectively. Equal Error Rate: 0.1867, Accuracy: 0.8132, Sensitivity: 0.7939, and Precision: 0.8362. Finally, Receive Operative Characteristic (ROC) gives 0.8034. Having fully explored and implemented this model, the performance of our model was examined and compared with another work and we found out that our model out-performed the model in terms of accuracy and precision.

Ogungbade BA