Development of a Bayesian belief network for the diagnosis of ventricular arrhythmias

Ventricular extrasystoles (VE) are considered the most dangerous type of heart rhythm disorders for human life, their timely detection, diagnosis and prevention urgent issues cardiology. In order to ensure objectivity VE, it is necessary process a large amount information related results various medical studies, tests, anamnesis, accompanying diseases, etc., along with long-term Holter ECG monitor. such make correct diagnosis, issue applying expert systems (MES) doctors currently relevant. ESs using probabilistic models based on Bayes' theorem preferred because there uncertainties in that same symptoms may be different diseases. The object this study development construction Bayesian belief network (BBN) purpose diagnosing VEs. choice BBN justified by fact they have ability combine several types information, as well manage work incomplete information. result application developed assessment VE. This was built NETICA system from Norsys Software Corp. A distinctive feature when compiling table conditional probabilities together echo-ECG data influence additional factors play role occurrence VE were used, index oxygen saturation erythrocytes blood, changes thickness intima-media layer aortic artery lipid fractions blood plasma