Data intelligence using PDME for predicting cardiovascular predictive failures

Abstract

In the area of Cardiovascular Diseases (CVD), dyspnea, one of many conditions that can be symptom of heart failure, is a metric used by New York Heart Association (NYHA) classification in order to describe the impact of heart failure on a patient. Based on four classes this classification measures the level of limitation during a simples physical activity. With the use of a non-invasive home tele monitoring system called Smart BEAT to retrieve biological data and heart metrics combined with a data-mining engine called PDME (Pervasive Data Mining Engine) is possible to obtain a different type of analysis sustained by a real time classification. The connection between the risk factors of CVD with the accuracy levels in the data models is recognizable, and continuously reflected with all the scenarios that were created. As soon, the data models used less CVD’s risk factors variables, the data models become useless, showing us how connected the risks are to this disease, this sustains the idea that PDME can be competent data mining engine in this field of work.