Arrhythmia Classification using Supervised Machine Learning Algorithms

Heart related disorders remain as one of the most chronic health issues in the world; generally, cardiovascular diseases are rated as a prime factor for morbidity and mortality worldwide. The early diagnosis and detection of such conditions may lead to imperative treatment measures and prevent unpleasant complications. This would considerably reduce the risk of serious health complications. One of the diagnostic tools for cardiac abnormalities is an electrocardiogram, a noninvasive, more generally adopted clinical investigative tool that captures over a period of time the heart's electrical function. Due to the fact that it is very accessible and easy to use, an ECG would normally be done first in diagnosing many cardiac conditions, including arrhythmias. The paper represents the detailed classification of arrhythmia, which is an irregularity in heart rhythm and can be a symptom of severe underlying cardiac problems based on ECG data. We have explored the efficacy of machine learning algorithms and Ensemble Method that combined the former for better predictions. Our results indicate that this ensemble learning approach performs better compared to all of the other individual algorithms used in this study, with an accuracy of 72.15%. The present study not only endorses the effort being made in better cardiac diagnostic tools but also calls for further attention toward advanced computational methods in health care to resolve some of the pivotal, complex challenges in medicine.