Atrial Fibrillation Discrimination for Real-Time ECG Monitoring Based On QT Interval Variation

Background/Objectives: An occasional Atrial Fibrillation (AF) event in heart rhythm should be monitored regularly, continuous intervals. Timely detection of these anomalies is required to save patients from sudden cardiac arrest. Method: A long-duration ECG categorization algorithm named AFECOC proposed. For this one-minute-long 71 signals are attained the Physionet’s “MIT-BIH arrhythmia (MA)” and “AF” database. Two-stage filtering noisy employed before signal analysis. Four algorithms i.e. Error-Correcting Output Code (ECOC), Naïve Bayes, Decision Tree, K-Nearest Neighbor(K-NN) applied reduce feature set then classified with ECOC classifier. Findings: It was found that gives highest accuracy 81.95% on complete set. To exclude irrelevant features, performing used extracts combination sets get most affected during AF. The ’heart-beat’ ’mean QT-interval’ relevant features AF events. two evaluated four classifiers namely ECOC, tree-based K mean classifier obtained 89.6%, 76.19%, 61% respectively. concludes proposed methodology achieved 89.6% Finally, all rhythms have been checked using annotated labels for spontaneous change QT-interval verify designed methodologies. Novelty: Instead missing P-waves RR-interval variation, recognition QT interval variation-based better longer signals. Hence, it can implemented Real- Time monitoring. Keywords: (AF); Classification; Error Correcting (ECOC); Feature extraction; QTinterval