Computer-Aided Atrial Fibrillation Diagnosis System with The Naive Bayesian Network: Based on The Analysis of 2016 Actual Cases of Electrocardiography Signals

Atrial Fibrillation (AF) is a form of arrhythmia that occurs often. Over one million people in the UK have been diagnosed with AF, which may result in the development of other severe disorders, eventually posing a risk to life. We developed a computer-aided approach for AF diagnosis based on the Naive Bayesian Network based on 2016 real electrocardiogram (ECG) signal cases in order to determine whether the candidate was healthy or has AF. Accuracy was up to 97%. Due to the simplicity of this technology, it potentially offers a low-cost solution for areas that cannot afford a costly AF diagnostic system. This paper details the whole process of the research, from its inception to its conclusion, including the first thoughts and the related work, the methodology, and the discussion. It also primarily demonstrates the program's technique, including the program's unique probabilistic reasoning process and the 10-fold cross-validation testing results.