Towards an Internet of Things application for the prognosis of Coronary Artery Disease using Machine Learning and Fuzzy Logic

Internet of Things (IoT) is a field of growing interest and reputation. It is a result of the advances of computer science and electronics, which enable the researchers and industries to design and create devices that connect with each other. In this work, we propose a novel application design for the prognosis of Coronary Artery Disease (CAD), which can exploit the capabilities of IoT devices for interacting with the user to gather data. We focus our work on the decision framework of the pipeline, which is implemented by Artificial Neural Networks, Decision Trees, and Fuzzy Cognitive Maps. The inner architecture of the model is powered by the real-time patient data. Its accuracy may be improved by taking advantage of heart rate signals, measured from wearable devices. The user is informed of the risk in several ways and is the one to monitor the functionality of the application through a graphical user interface (GUI). The simulation results demonstrate the effective functionality of the decision-making framework, since it manages to aggregate conflicting predictions and improve the accuracy. The proposed framework could be utilized for evaluation tests with the appropriate supplement of data.