A review on the performance of classification and prediction algorithms on cardiology data for the prediction of treadmill test through a mobile application

摘要

The availability of data related to heart diseases has increased enormously over the past several decades. This exponential growth of data necessitates the use of powerful data analysis tools to discover 'hidden' patterns in the medical database for effective decision making. Data mining techniques have been widely used in medical research, particularly in heart disease prediction, because of their ability to extract 'useful' knowledge and relationships from unstructured or semi-structured data. At the same time, digitisation of medical information and rapid usage of smart phones have led to the development of a large variety of mobile apps for disease diagnosis. In the field of cardiology, as treadmill test (TMT) has been a challenging procedure, a smart mobile device can be employed for the self-assessment of the test. Our research has led to the development of a mobile application for the prediction of the result of a treadmill test (TMT) utilising minimal number of clinical attributes using data mining algorithms.