Reаl-time heаrt diseаse detection аnd monitoring system bаsed on fаst mаchine leаrning using Spаrk

Abstract

Heart disease is one of the most prominent and dangerous diseases that threaten public health around the world, often leading to heart attacks and strokes. More recently, the Internet of Things (IoT) based health monitoring system and many others continuously generate enormous amounts of data. With the heart disease patients under supervision have been increasing significantly, data generation speed is very high. In the scope of application of machine learning to the health data, there is a significant need to real-time streaming data processing to ensure an effective and scalable solution to effectively find and prevent the heart disease within a short and very specific timeline. As the amount of available data becomes highly big, the process of building and testing the traditional machine learning can be quite time consuming. The combination of streaming big data and machine learning is a breakthrough technology that can have a significant impact in the healthcare field especially real-time detection of heart disease. In this paper, we propose a new heart disease monitoring system based on a new classification approach consisting of the real-time distributed machine learning which uses the real-time predictive analysis algorithm in the Spark environment to predict heart disease. Firstly, we transform the traditional decision tree (C4.5) algorithm into parallel, distributed, scalable and fast decision tree. Secondly, this model is applied to streaming data coming from distributed sources to predict heart disease in real-time. The result as well as data streams will be stored in a distributed database for real-time reporting and monitoring. The results show better performance and have the potential to open new opportunities in distributed machine learning.