Classification of Arrhythmia Disease using Enhanced RNN Model

  • B.Bavani, S. Nirmala Sugirtha Rajini, M.S. Josephine, V.Prasannakumari
Keywords: Heartbeat, Arrhythmia, Prediction, Deep Learning, Decision Tree, Recurrent Neural Network

Abstract

The main cause for heart disease is high blood pressure, smoking, high blood cholesterol. Various other medical states and lifestyle choices can also put people at a higher risk for heart disease, including obesity, overweight, and diabetes. Most of the health issues are unhealthy lifestyles like inadequate exercise, bad diet, and smoking. Heart disease is of various types. Arrhythmia is called a rhythm abnormality. The wave line of ECG is in minute size, so the detection of Arrhythmia through the naked eyes is a challenging task for those who are in the health care field. For this reason, deep learning concepts are applied for the detection of the disease. In this paper, the Enhanced RNN model is being applied for finding the abnormality of the heartbeat. The proposed system's outcome is compared with the Decision Tree (DT) and RNN(Recurrent Neural Network) model. Comparing this enhanced RNN yields better results based on accuracy, precision, and recall. The proposed system produces 91% accuracy, 90% precision, and 93% recall.  The heartbeat data is gathered from the online dataset Kaggle.  The proposed system is implemented using Python programming. It consists of twelve kinds of attributes, including patient’s health data and ECG data.

Published
2021-07-21
How to Cite
M.S. Josephine, V.Prasannakumari, B. S. N. S. R. (2021). Classification of Arrhythmia Disease using Enhanced RNN Model. Design Engineering, 4062- 4072. Retrieved from http://www.thedesignengineering.com/index.php/DE/article/view/2842
Section
Articles