Authors: Kartik Bhanot; Reshub Kr. Nigam
Addresses: Information and Communication Technology Department, Manipal Institute of Technology, Manipal University, Karnataka-576104, India ' Information and Communication Technology Department, Manipal Institute of Technology, Manipal University, Karnataka-576104, India
Abstract: Cardiovascular diseases are the leading cause of death all across the globe, much more than all forms of cancer combined. In order to study these heart ailments and stress levels of patients, electrocardiogram (ECG) data is used. The challenge is to develop a risk analysis model that can determine the risk or the possibility of a heart attack based on the current state of ECG data. In the current paper, authors have developed a deterministic risk analysis to determine the level of risk that a person may have for a heart attack with an average accuracy of 92.57%. Neural networks have been used extensively for training the developed model for analysis purposes. The data has been taken from (Massachusetts Institute of Technology - Boston's Beth Israel Hospital) MIT-BIH Long-term ECG Database and MIT-BIH Arrhythmia Database using Rapid Miner as the platform. The database is divided into equitable data from long-term ECG as well as from arrhythmia patients and labels is assigned to them so as to maintain the legitimacy of the whole dataset.
Keywords: electrocardiograms; ECG signals; neural networks; RapidMiner; smart sensors; e-healthcare; electronic healthcare; QRS complex; data analysis; decision tree; risk assessment; heart attacks; stress levels; heartrate variables; Pam Tomkins technique; smart health models; cardiovascular disease; arrhythmia.
International Journal of Telemedicine and Clinical Practices, 2017 Vol.2 No.1, pp.63 - 73
Received: 15 Jun 2016
Accepted: 18 Jul 2016
Published online: 07 Feb 2017 *