Detection of Atrial Fibrillation in ECGs

Automatic detection and classification of arrhythmia in electrocardiograms (ECG) provides a framework for efficient diagnosis and broader outreach to patients at risk for cardiac diseases. While prevalent types of arrhythmia include premature ventricular contractions (PVC) and atrial fibrillation, the majority of existing literature focuses on automatic detection and classification of the former type. In this report, we discuss our heuristic and implementation for detection of atrial fibrillation, using a dataset provided by the MIT/BIH Arrhythmia Database. Using features from Fourier analysis, wavelet transformation, and R-R interval analysis, linear discriminant analysis (LDA) on individual segments performed well with classification error of approximately 10%. Our detector, which built on top of our classifier, successfully identified regions of atrial fibrillation with less than 2% error.