Towards IoT and ML Driven Cardiac Status Prediction System

The amount of death rate in Bangladesh because of different types of non-communicable cardiac diseases are rapidly increasing every year. These non-communicable diseases are long-term and slowly progressive and become so severe that the condition of a patient is too critical to handle. As a result, most of the people getting a sudden heart attack or learn about their disease when it is too late. Thus, in this paper, we propose an IoT and Machine Learning derived cardiac status prediction method that can collect necessary data from human body using IoT devices (sensors) and transfer those data to the cloud where the data are stored with user authentication. The data which are received from the human body are then to be normalized before machine learning algorithms are applied to them to calculate and predict the overall condition of a patients heart. To-date we could collect heart rate, ECG signal, and cholesterol through IoT devices from the human body. The results we have received of heart rate and cholesterol are quite satisfactory. There are still some issues with the interpretation of the ECG signal which we are working on.